Method for estimating wind uplift resistance static strength of metal roof

Abstract

The invention relates to the technical field of metal roof strength estimation, in particular to a metal roof wind uplift resistance static strength estimation method which comprises the following steps: S1, building a metal roof wind uplift resistance test sample; s2, testing the static wind uplift resistance of the test sample; s3, collecting data related to the wind uplift strength of the metal roof; s4, carrying out numerical processing; s5, establishing a hierarchical support vector regression model; and S6, performing real-time monitoring and maintenance. According to the method for estimating the wind uplift resistance static strength of the metal roof, actual wind uplift resistance information of the roof is obtained by building a test sample, carrying out a static wind uplift resistance test, collecting related data and the like; meanwhile, a hierarchical support vector regression model is established by using the data so as to realize the combination of wind uplift resistance static strength prediction, physical data testing and an algorithm model and ensure that the metal roof has enough wind uplift resistance in practical application.

Description

Technical Field

[0001] The invention relates to the technical field of metal roof strength estimation, and in particular to a method for estimating the static strength of a metal roof against wind uplift. Background Art

[0002] Traditional methods for evaluating the static strength of metal roofs against wind uplift mainly rely on physical tests and empirical formulas. Although these methods can meet engineering needs to a certain extent, they have many limitations. Physical tests are costly, time-consuming and labor-intensive, and it is difficult to simulate complex wind conditions and roof structural characteristics. Empirical formulas are often based on simplified assumptions and limited experimental data, and their accuracy and versatility are limited. With the rapid development of computer technology and numerical analysis methods, people have begun to try to use numerical simulation to predict the wind-uplift resistance of metal roofs. By establishing a sophisticated mathematical model, the interaction between wind and roof can be more accurately described, thereby providing a more reliable basis for estimating the wind-uplift strength. However, numerical simulation methods require a lot of computing resources and expertise, and the accuracy of the model still needs to be further verified.

[0003] In recent years, the emergence of machine learning technology has brought new opportunities for the estimation of static strength of metal roofs against wind uplift. Machine learning algorithms can automatically learn hidden patterns and laws from a large amount of data, thereby establishing a more accurate prediction model. By combining actual monitoring data, numerical simulation results, and physical test data, machine learning models can comprehensively consider the influence of multiple factors and improve the accuracy and reliability of wind uplift strength estimation. Summary of the invention

[0004] The technical problem to be solved by the present invention is: in order to solve the technical problems in the prior art, the present invention provides a method for estimating the static strength of a metal roof against wind uplift.

[0005] The technical solution adopted by the present invention to solve the technical problem is: a method for estimating the static wind-uplift resistance strength of a metal roof, comprising the following steps: S1, building a metal roof wind-uplift resistance test sample; S2, testing the static wind-uplift resistance of the test sample, loading different pressures on the test sample, pressurizing step by step each time, checking whether the roof system is damaged or failed, and recording the pressure value of the previous level of damage or failure as the wind-uplift resistance pressure; S3, collecting data related to the wind-uplift resistance strength of the metal roof, including the roof shape, slope, plate thickness, connection method, wind speed of the wind load conditions of the tested sample, and the wind speed of the wind load conditions; S4, collecting data related to the wind-uplift resistance strength of the metal roof, including the roof shape, slope, plate thickness, connection method, wind speed of the wind load conditions of the tested sample, and wind speed of the wind load conditions of the tested sample. Speed, wind direction, and turbulence; S4, numerical processing, numerical processing of the strength, elastic modulus, roof shape, and slope parameters of the metal sheet; S5, establishment of a hierarchical support vector regression model, decomposition of the static strength prediction problem of wind resistance into multiple levels, each level uses support vector regression for local regression, and then combines the results of each level to obtain the final wind resistance static strength prediction result; S6, real-time monitoring and maintenance, by using stress sensors to detect the stress condition of the metal roof, and taking corresponding measures in a timely manner according to the detection results and prediction results.

[0006] The static strength estimation method of a metal roof against wind uplift of the present invention can ensure that the wind uplift resistance of the metal roof in actual application meets the design requirements through the construction of test samples and the static wind uplift resistance test. At the same time, the data of the test samples are collected and numerically processed, providing detailed data support for the metal roof design. These data help engineers to more accurately evaluate the influence of different design parameters on the wind uplift resistance, thereby optimizing the design parameters such as the roof shape, slope, plate thickness and connection method, so as to achieve better wind uplift resistance effect and economy. At the same time, this systematic testing process helps to identify and improve the weak links in the roof structure, thereby improving the reliability and safety of the overall structure.

[0007] Furthermore, in step S1, the test sample installation steps include purlin installation, corrugated steel base plate installation, leveling steel plate installation, waterproof layer laying, noise reduction layer laying, metal roof panel installation, and wind-resistant clip installation.

[0008] Furthermore, in step S1, the test sample is placed in a test box, and the test sample is connected to the bottom of the test box, and the test box is closed.

[0009] Furthermore, the test box is connected to an air inlet pipe, the air inlet pipe is provided with an air inlet valve, the air inlet pipe is arranged below the metal roof panel of the test sample, the top of the test box is connected to an air outlet pipe, and the air outlet pipe is connected to an induced draft fan.

[0010] Furthermore, in step S2, different pressures are applied to the test sample through the induced draft fan and the air inlet valve, and a stable pressure is applied to the metal roof for 60 seconds.

[0011] Furthermore, a pressure sensor is installed in the test box, and the pressure sensor is used to detect the metal roof.

[0012] Furthermore, step S4 also includes the following steps:

[0013] S41. Numerical processing of the strength and elastic modulus of metal sheets. The strength indicators of metal sheets include tensile strength, yield strength, and shear strength, which are measured in megapascals, and the elastic modulus is measured in MPa.

[0014] S42, the roof shape is described by geometric parameters, and the roof slope is numerically expressed by angle;

[0015] S43. Wind speed is quantified in meters per second, wind direction is quantified in angles, and turbulence intensity is quantified in dimensionless parameters.

[0016] Furthermore, in step S42, the roof shape described by non-geometric parameters is described using curvature and curvature radius.

[0017] Furthermore, step S5 further includes the following steps:

[0018] S51, the first layer SVR, takes the data processed by numerical processing in S4 as input variable x, and divides the input variable into K groups, that is, x = (x (1) ,x (2) ,…,x (K) ), in the first layer, for each group's input variable x (k) Use SVR for regression and get the prediction result: y (k) =f (k) (x (k) ),

[0019] Among them, f (k) is the prediction function of the kth SVR model, y (k) is the prediction result of the kth SVR model. For each SVR model, the following optimization problem is used to solve it:

[0020]

[0021] subject to

[0022]

[0023] Among them, w (k) and b (k) are the coefficients of the kth SVR model, and is a slack variable used to deal with noise and outliers in the data, C is a penalty coefficient, and ε is an insensitive loss parameter. The input variable Function mapped to high-dimensional feature space, i is the index of the sample, N represents the total number of training samples, is the prediction result of the kth SVR model for the sample;

[0024] S52, the second layer and above SVR, the prediction result y of the first layer (k) As the input variable, SVR is used for regression to obtain a higher-level prediction result. M SVR models are set in the second layer, and the prediction result of the mth SVR model is expressed as:

[0025] y (m) =f (m) (y (1) ,y (2) ,…,y (K) )

[0026] The parameter solving problem is the same as step S51;

[0027] S53, prediction results, through progressive regression, the final prediction results are:

[0028] y=f(y (1) ,y (2) ,...,y (M) )

[0029] Among them, f is the final prediction function and y is the final prediction result.

[0030] Furthermore, step S6 also includes the following steps:

[0031] S61, the stress sensor monitors the stress condition of the metal roof in real time during actual use, and transmits the detected stress data to the signal processing system for analysis, and determines whether the current stress condition of the metal roof is normal by analyzing the detection results;

[0032] S62, comparing the detection result with the prediction result of the hierarchical support vector regression model established in S5, and if there is a large deviation between the detection result and the prediction result, further optimizing the model;

[0033] S63. Based on the analysis of the test results and prediction results, strengthen the weak points to ensure that the connections between the various components are firm and reliable.

[0034] The beneficial effects of the present invention are:

[0035] 1. Through the construction of test samples and static wind-resistant uplift test, it can be ensured that the wind-resistant performance of metal roofs in actual applications meets the design requirements. At the same time, the data of the test samples are collected and numerically processed, providing detailed data support for the design of metal roofs. These data help engineers to more accurately evaluate the impact of different design parameters on the wind-resistant strength, thereby optimizing the design parameters such as roof shape, slope, sheet thickness and connection method to achieve better wind-resistant effect and economy; use these data to establish a hierarchical support vector regression model to predict the static strength of wind-resistant uplift. This systematic testing process helps to identify and improve the weak links in the roof structure, thereby improving the reliability and safety of the overall structure.

[0036] 2. During the experiment, the algorithm model can learn hidden patterns and laws from a large amount of data, thereby improving the accuracy and reliability of wind-resistant strength estimation; by comparing and analyzing the model's prediction results with the actual test results, the model can be further optimized to improve its prediction ability; ensuring that the metal roof has sufficient wind-resistant ability in actual applications, the combination of physical data testing and algorithm models is a comprehensive means to achieve this goal. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0038] Figure 1 The present invention is a schematic flow chart of a method for estimating the static strength of a metal roof against wind uplift.

[0039] Figure 2 It is a schematic diagram of the installation of the test sample and the test box in the present invention.

[0040] Figure 3 It is a schematic diagram of the test sample embodied in the present invention.

[0041] Figure 4 It is a schematic diagram of the entire test box in the present invention.

[0042] Figure 5 It is a schematic diagram of the mounting bracket in the present invention.

[0043] Figure 6 It is a schematic diagram showing the position of the vent plug in the present invention.

[0044] Figure 7 It is a graph of the network prediction results in step S5 of the present invention.

[0045] In the figure: 1. test box; 2. test sample; 21. purlin; 22. corrugated steel bottom plate; 23. leveling steel plate; 24. waterproof layer; 25. noise reduction layer; 26. metal roof panel; 27. wind-resistant clip; 3. air inlet valve; 4. induced draft fan; 5. pressure sensor; 6. transparent observation window; 7. mounting bracket; 71. main pipe; 72. horizontal pipe; 73. vertical pipe; 74. rotating rod; 741. hand wheel; 75. screw; 76. synchronous plate; 77. sealing plug. DETAILED DESCRIPTION

[0046] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0047] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "multiple" means two or more. In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be a connection between the two elements. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0048] The invention discloses a method for estimating the static strength of a metal roof against wind uplift.

[0049] Reference Figures 1 to 6 A method for estimating the static strength of a metal roof against wind uplift comprises the following steps:

[0050] S1. Construct the metal roof wind-resistant uplift test sample 2. The test sample 2 is installed in the test box 1. The test sample 2 is connected to the bottom of the test box 1, and the test box 1 is closed to reduce interference factors.

[0051] The test sample 2 includes a purlin 21, a corrugated steel base plate 22, a leveling steel plate 23, a waterproof layer 24, a noise reduction layer 25, a metal roof panel 26 and a wind-resistant clip 27. During installation, the installation steps include installing the purlin 21, installing the corrugated steel base plate 22, installing the leveling steel plate 23, laying the waterproof layer 24, laying the noise reduction layer 25, installing the metal roof panel 26, and installing the wind-resistant clip 27.

[0052] Specifically, the test box 1 is connected with an air inlet pipe, on which an air inlet valve 3 is provided, the air inlet pipe is arranged below the metal roof panel of the test sample 2, the top of the test box 1 is connected with an air outlet pipe, and the air outlet pipe is connected with an induced draft fan 4. A pressure sensor 5 is installed in the test box 1, and the pressure sensor 5 is used to detect the metal roof. A transparent observation window 6 is also installed on the test box 1 to observe the test situation.

[0053] The fan is installed on the top of the test box 1 through the mounting bracket 7. The air outlet duct includes a main pipe 71, multiple transverse pipes 72 and multiple vertical pipes 73. The multiple vertical pipes 73 are all connected to the top of the test box 1. The multiple vertical pipes 73 are connected through multiple transverse pipes 72 so that the multiple vertical pipes 73 are connected to the main pipe 71. The top of the main pipe 71 is connected to the fan, the bottom of the main pipe 71 is connected to the test box 1, and the side wall of the main pipe 71 is connected to the transverse pipe 72. In this embodiment, the main pipe 71 is located in the middle of the top surface of the test box 1. The vertical pipes 73 can be two or four, and in this embodiment, there are six, which are distributed around the main pipe 71, so as to facilitate the uniformity of air suction. In this embodiment, the induced draft fan 4 is used to achieve air suction, and in another embodiment, the induced draft fan 4 can also be used for blowing air.

[0054] Two rotating rods 74 are rotatably connected to the mounting bracket 7, and a screw rod 75 is coaxially fixed to each rotating rod 74. The two screw rods 75 are threadedly connected to the same synchronous plate 76, and six sealing plugs 77 are fixedly connected to the synchronous plate 76. The six sealing plugs 77 correspond to the six vertical pipes 73 one by one, and the top of each vertical pipe 73 is penetrated so that the sealing plug 77 can be inserted into the vertical pipe 73. The sealing plug 77 is adapted to the vertical pipe 73, so that the sealing plug 77 can block the junction between the vertical pipe 73 and the horizontal pipe 72, and the height of the sealing plug 77 can be adjusted to adjust the area of ​​the connection port at the junction of the horizontal pipe 72 and the vertical pipe 73, thereby adjusting the flow rate of the vertical pipe 73 entering the horizontal pipe 72, so as to better apply pressures of different sizes and ranges to the test sample 2.

[0055] In addition, a hand wheel 741 is fixedly connected to the rotating rod 74 to facilitate the rotation of the rotating rod 74. When the flow rate of the vertical pipe 73 entering the horizontal pipe 72 needs to be adjusted, the hand wheel 741 is rotated to drive the rotating rod 74 to rotate, and the rotating rod 74 drives the screw 75 to rotate, thereby driving the synchronous plate 76 to move in the vertical direction, so as to adjust the height of the sealing plug 77. The sealing plug 77 can completely block the connection between the horizontal pipe 72 and the vertical pipe 73, so that the gas only passes through the main pipe 71.

[0056] S2. Static wind-uplift resistance test of test sample 2. Different pressures are applied to test sample 2 through induced draft fan 4 and air inlet valve 3 during pressurization. Stable pressure is applied to the metal roof for 60 seconds. The pressure is increased step by step each time to check whether the roof system is damaged or failed. The pressure value of the previous level of damage or failure is recorded as the wind-uplift resistance pressure.

[0057] S3. Collect data related to the wind-uplift resistance of metal roofs, including the roof shape, slope, sheet thickness, connection method, wind speed, wind direction, and turbulence of the tested sample 2. Use laser rangefinders and angle measuring instruments to obtain information on the shape and slope of the roof; use calipers to measure the sheet at multiple points and collect thickness data; understand the connection method between the various components of the roof through on-site observation and recording, including the type, quantity, and location of fasteners; install anemometers, wind vanes, and turbulence measurement equipment near the roof to monitor wind-related data in real time to obtain wind speed, wind direction, and turbulence under wind load conditions.

[0058] S4. Numerical processing: numerical processing of the strength, elastic modulus, roof shape and slope parameters of the metal sheet.

[0059] The digitization process also includes the following steps:

[0060] S41. Numerical processing of the strength and elastic modulus of metal sheets. The strength indicators of metal sheets include tensile strength, yield strength, and shear strength, which are measured in megapascals, and the elastic modulus is measured in MPa.

[0061] S42. The roof shape is described by geometric parameters, such as circle, square, triangle, etc. The roof slope is numerically expressed by angle. For roof shapes that are difficult to describe by geometric parameters, curvature, curvature radius, etc. are used to describe them;

[0062] S43, wind speed is quantified in meters per second, wind direction is quantified in angles, and turbulence intensity is quantified in dimensionless parameters;

[0063] S5. Establish a hierarchical support vector regression model to decompose the static strength prediction problem of wind-resistant uplift into multiple levels. Use support vector regression for local regression at each level, and then combine the results of each level to obtain the final static strength prediction result of wind-resistant uplift.

[0064] Building a hierarchical support vector regression model also includes the following steps:

[0065] S51, the first layer SVR, takes the data processed by numerical processing in S4 as input variable x, and divides the input variable into K groups, that is, x = (x (1),x (2) ,...,x (K) ), in the first layer, SVR is used to regress the input variables of each group and obtain the prediction results: y (k) =f (k) (x (k) ),

[0066] Among them, f (k) is the prediction function of the kth SVR model, y (k) is the prediction result of the kth SVR model. For each SVR model, the following optimization problem is used to solve it:

[0067]

[0068] subject to

[0069]

[0070] Among them, w (k) and b (k) are the coefficients of the kth SVR model, and is a slack variable used to deal with noise and outliers in the data, C is a penalty coefficient, and ε is an insensitive loss parameter. The input variable Function mapped to high-dimensional feature space, i is the index of the sample, N represents the total number of training samples, is the prediction result of the kth SVR model for the sample;

[0071] S52, the second layer and above SVR, the prediction result y of the first layer (k) As the input variable, SVR is used for regression to obtain a higher-level prediction result. M SVR models are set in the second layer, and the prediction result of the mth SVR model is expressed as:

[0072] y (m) =f (m) (y (1) ,y (2) ,...,y (K) )

[0073] The parameter solving problem is the same as step S51;

[0074] S53, prediction results, through progressive regression, the final prediction results are:

[0075] y=f(y (1) ,y (2) ,...,y (M) )

[0076] Among them, f is the final prediction function, y is the final prediction result, refer to Figure 7 .

[0077] S6. Real-time monitoring and maintenance: Use stress sensors to detect the stress conditions of metal roofs and take appropriate measures in a timely manner based on the test results and prediction results.

[0078] Real-time monitoring and maintenance also includes the following steps:

[0079] S61. The stress sensor monitors the stress condition of the metal roof in real time during actual use. The stress sensor transmits the detected stress data to the signal processing system for analysis. By analyzing the detection results, it is determined whether the current stress state of the metal roof is normal. If it is found that the stress exceeds the threshold or abnormal stress distribution occurs, it may indicate that there is a problem with the roof's wind-resistant ability. In this case, the roof needs to be strengthened.

[0080] S62, compare the test results with the prediction results of the hierarchical support vector regression model established in S5. If there is a large deviation between the test results and the prediction results, further optimize the model. The prediction results of the model can provide a reference benchmark for the test results. By comparing the test results with the prediction results, it can be determined whether the test results are within a reasonable range. After the roof panel is damaged, compare the test results with the prediction results. If the test results differ greatly from the prediction results, it may indicate that there are problems in the test process or the actual situation of the roof has changed, and the model needs to be further optimized.

[0081] S63. Based on the analysis of the test results and prediction results, strengthen the weak points to ensure that the connections between the various components are firm and reliable.

[0082] Based on the above ideal embodiments of the present invention, the relevant staff can make various changes and modifications without departing from the technical concept of the present invention through the above description. The technical scope of the present invention is not limited to the contents of the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A method for estimating the static strength of metal roofs against wind uplift, characterized in that: The steps include: S1. Construct metal roof wind-resistant uplift test samples (2); S2. Static wind-uplift resistance test of the test sample (2): different pressures are applied to the test sample (2), and the pressure is increased step by step each time to check whether the roof system is damaged or failed. The pressure value before damage or failure is recorded as the wind-uplift resistance pressure; S3, collecting data related to the wind-uplift resistance of metal roofs, including the roof shape, slope, plate thickness, connection method, wind speed, wind direction, and turbulence of the tested sample (2); S4, numerical processing, numerical processing of the strength, elastic modulus, roof shape, and slope parameters of the metal sheet; S5. Establish a hierarchical support vector regression model to decompose the static strength prediction problem of wind-resistant uplift into multiple levels. Use support vector regression to perform local regression at each level, and then combine the results of each level to obtain the final static strength prediction result of wind-resistant uplift; S6. Real-time monitoring and maintenance: Use stress sensors to detect the stress conditions of metal roofs and take appropriate measures in a timely manner based on the test results and prediction results.

2. A method for estimating the static strength of metal roofing against wind uplift according to claim 1, characterized in that: In step S1, the installation steps of the test sample (2) include installation of purlins, installation of corrugated steel base plate, installation of leveling steel plate, laying of waterproof layer, laying of noise reduction layer, installation of metal roof panel, and installation of wind-resistant clips.

3. A method for estimating the static strength of metal roofing against wind uplift according to claim 2, characterized in that: In step S1, the test sample (2) is placed in the test box (1), and the test sample (2) is connected to the bottom of the test box (1), and the test box (1) is closed.

4. A method for estimating the static strength of metal roofing against wind uplift according to claim 3, characterized in that: The test box (1) is connected to an air inlet pipe, on which an air inlet valve (3) is provided. The air inlet pipe is arranged below the metal roof panel of the test sample (2). The top of the test box (1) is connected to an air outlet pipe, and the air outlet pipe is connected to an induced draft fan (4).

5. A method for estimating the static strength of metal roofing against wind uplift as claimed in claim 4, characterized in that: In step S2, different pressures are applied to the test sample (2) through the induced draft fan (4) and the air inlet valve (3), and a stable pressure is applied to the metal roof for 60 seconds.

6. A method for estimating the static strength of a metal roof against wind uplift as claimed in claim 5, characterized in that: A pressure sensor (5) is installed in the test box (1), and the pressure sensor (5) is used to detect the metal roof.

7. A method for estimating the static strength of metal roofing against wind uplift as claimed in claim 1, characterized in that: The S4 step also includes the following steps: S41. Numerical processing of the strength and elastic modulus of metal sheets. The strength indicators of metal sheets include tensile strength, yield strength, and shear strength, which are measured in megapascals, and the elastic modulus is measured in MPa. S42, the roof shape is described by geometric parameters, and the roof slope is numerically expressed by angle; S43. Wind speed is quantified in meters per second, wind direction is quantified in angles, and turbulence intensity is quantified in dimensionless parameters.

8. A method for estimating the static strength of a metal roof against wind uplift as claimed in claim 7, characterized in that: In step S42, the roof shape described by non-geometric parameters is described using curvature and curvature radius.

9. A method for estimating the static strength of metal roofing against wind uplift as claimed in claim 1, characterized in that: The step S5 also includes the following steps: S51, the first layer SVR, takes the data processed by numerical processing in S4 as input variable x, and divides the input variable into K groups, that is, x = (x (1) ,x (2) ,…,x (K) ), in the first layer, for each group's input variable x (k) Use SVR for regression and get the prediction result: y (k) =f (k) (x (k) ), Among them, f (k) is the prediction function of the kth SVR model, y (k) is the prediction result of the kth SVR model. For each SVR model, the following optimization problem is used to solve it: subject to Among them, w (k) and b (k) are the coefficients of the kth SVR model, and is a slack variable used to deal with noise and outliers in the data, C is a penalty coefficient, and ε is an insensitive loss parameter. The input variable Function mapped to high-dimensional feature space, i is the index of the sample, N represents the total number of training samples, is the prediction result of the kth SVR model for the sample; S52, the second layer and above SVR, the prediction result y of the first layer (k) As the input variable, SVR is used for regression to obtain a higher-level prediction result. M SVR models are set in the second layer, and the prediction result of the mth SVR model is expressed as: and (m) =f (m) (and (1) ,and (2) ,…,and (K) ) The parameter solving problem is the same as step S51; S53, prediction results, through progressive regression, the final prediction results are: y=f(y (1) ,and (2) ,…,and (M) ) Among them, f is the final prediction function and y is the final prediction result.

10. A method for estimating the static strength of a metal roof against wind uplift as claimed in claim 6, characterized in that: Step S6 also includes the following steps: S61, the stress sensor monitors the stress condition of the metal roof in real time during actual use, and transmits the detected stress data to the signal processing system for analysis, and determines whether the current stress condition of the metal roof is normal by analyzing the detection results; S62, comparing the detection result with the prediction result of the hierarchical support vector regression model established in S5, and if there is a large deviation between the detection result and the prediction result, further optimizing the model; S63. Based on the analysis of the test results and prediction results, strengthen the weak points to ensure that the connections between the various components are firm and reliable.

Images