Point-supported glass curtain wall damage degree quantitative determination method

Abstract

The present application relates to the technical field of building curtain wall safety detection, and particularly relates to a point-supported glass curtain wall damage degree quantitative determination method, which utilizes a complete same replica point-supported glass curtain wall as the to-be-detected point-supported glass curtain wall, obtains a function relationship between a damage degree identification index of a preset position of a glass panel of the replica point-supported glass curtain wall and a bolt pretightening force of the glass panel of the replica point-supported glass curtain wall, calculates the damage degree identification index of the preset position of the glass panel of the to-be-detected point-supported glass curtain wall, and utilizes the function relationship to calculate a current bolt pretightening force of the glass panel of the to-be-detected point-supported glass curtain wall, so as to realize quantitative identification of the bolt pretightening force, thereby achieving the effect of quantitatively judging the damage degree of the point-supported glass curtain wall. The method has strong operability, high reliability, and is suitable for on-site large-volume curtain wall detection and subsequent repair work guidance.

Description

Technical Field

[0001] This invention relates to the field of building curtain wall safety inspection technology, and in particular to a method for quantitatively determining the degree of damage to point-supported glass curtain walls. Background Technology

[0002] Due to their aesthetically pleasing and ornate artistic form, building curtain walls have been widely used in my country since their introduction in the 1980s. According to statistics from China Curtain Wall Network, the equivalent of existing building curtain walls in my country exceeds 1.5 billion square meters. However, with the widespread use and increasing service life of building curtain walls, safety issues have gradually emerged, becoming a significant problem affecting people's livelihoods. Among various types of building curtain walls, frameless glass curtain walls, which rely solely on structural adhesive connections, have been gradually abandoned due to their unreliable connection method, while the market for more secure and reliable point-supported glass curtain walls is rapidly expanding.

[0003] In the field of curtain wall safety inspection and monitoring, research on vibration-based detection methods is relatively advanced. Existing technologies include methods for judging the damage level of frameless glass curtain walls based on the cumulative difference of the frequency response function at the origin; methods that use laser vibrometers to collect the first-order natural frequency of the frameless glass curtain wall and use it as an indicator to judge the degree of damage; and methods that use the cumulative change in vibration transmissibility to judge the damage level of frameless glass curtain walls. However, most methods only make simple comparisons between intact and damaged conditions using damage indicators, failing to provide a precise quantitative evaluation of the actual damage level of the curtain wall. This is detrimental to the continued use of curtain walls with minor damage, and also fails to distinguish between curtain walls that can continue to be used after repair and those that must be completely replaced. Furthermore, relatively few safety inspection technologies for point-supported glass curtain walls have been disclosed. Therefore, there is an urgent need for a technical method that can quantitatively determine the damage level of point-supported glass curtain walls. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a quantitative method for determining the degree of damage to point-supported glass curtain walls, which addresses the shortcomings of the prior art.

[0005] The technical solution of the present invention for a method for quantitatively determining the degree of damage to a point-supported glass curtain wall is as follows:

[0006] Using a replica point-supported glass curtain wall that is exactly the same as the one to be tested, the functional relationship between the damage degree identification index of the glass panel at a preset position of the replica point-supported glass curtain wall and the bolt preload of the glass panel of the replica point-supported glass curtain wall is obtained.

[0007] Calculate the damage degree identification index of the glass panel at a preset position of the point-supported glass curtain wall to be tested, and use the functional relationship to calculate the current bolt preload of the glass panel of the point-supported glass curtain wall to be tested, wherein the preset position of the glass panel of the point-supported glass curtain wall to be tested relative to the glass panel of the point-supported glass curtain wall to be tested is the same as the preset position of the glass panel of the composite point-supported glass curtain wall relative to the glass panel of the composite point-supported glass curtain wall.

[0008] The beneficial effects of the method for quantitatively determining the degree of damage to point-supported glass curtain walls according to the present invention are as follows:

[0009] 1) The method proposed in this invention can realize the quantitative identification of the pre-tightening force of bolts in point-supported glass curtain walls, thereby achieving the effect of quantitatively judging the degree of damage. It is highly operable and reliable, and can provide technical support for curtain wall inspection and maintenance.

[0010] 2) The method proposed in this invention comprehensively considers the changing trends of multiple natural frequencies and centroid frequencies under different damage levels of point-supported glass curtain walls, and performs index fusion and quantitative processing, which greatly improves the detection accuracy.

[0011] 3) This invention utilizes laser vibration measurement technology, which enables remote, non-contact, and non-destructive testing of point-supported glass curtain walls, greatly improving testing efficiency and making it suitable for large-scale on-site curtain wall testing. Attached Figure Description

[0012] Figure 1 This is a schematic flowchart of a method for quantitatively determining the degree of damage to a point-supported glass curtain wall provided in an embodiment of the present invention;

[0013] Figure 2 This is a schematic diagram of a point-supported glass curtain wall structure in an embodiment of the present invention;

[0014] Figure 3 This is the vibration signal spectrum diagram of working condition 9 in the embodiment of the present invention;

[0015] Figure 4 This is a comparison diagram of higher-order and lower-order natural frequencies in an embodiment of the present invention;

[0016] Figure 5 This is a graph showing the trend of vibration signal power spectrum as damage intensifies in an embodiment of the present invention;

[0017] Figure 6 This is a graph showing the change of the power spectrum centroid frequency with bolt preload in an embodiment of the present invention.

[0018] Figure 7 f is the embodiment of the present invention. ED Comparison chart of recognition accuracy with first-order natural frequency.

[0019] Figure 8 The damage degree identification index f in the embodiments of the present invention ED The curve of the fitting function with the bolt preload x. Detailed Implementation

[0020] The following description, in conjunction with specific implementation methods, provides an explanation.

[0021] Example 1:

[0022] like Figure 1 As shown in the figure, a method for quantitatively determining the degree of damage to a point-supported glass curtain wall according to an embodiment of the present invention includes the following steps:

[0023] S1. Using a replica point-supported glass curtain wall that is exactly the same as the one to be tested, obtain the damage identification index of the glass panel at the preset position of the replica point-supported glass curtain wall and the functional relationship between the bolt preload of the glass panel of the replica point-supported glass curtain wall.

[0024] S2. Calculate the damage identification index of the preset position of the glass panel of the point-supported glass curtain wall to be tested, and use the functional relationship to calculate the current bolt preload of the glass panel of the point-supported glass curtain wall to be tested. The preset position of the glass panel of the point-supported glass curtain wall to be tested relative to the position of the glass panel of the point-supported glass curtain wall to be tested is the same as the preset position of the glass panel of the composite point-supported glass curtain wall relative to the position of the glass panel of the composite point-supported glass curtain wall.

[0025] Optionally, in the above technical solution, the process of obtaining the functional relationship in S1 includes:

[0026] S10. Collect and calculate the vibration data of the glass panel of the composite point-supported glass curtain wall under different bolt preloads based on the preset position of the glass panel, calculate the centroid frequency of the vibration signal power spectrum, and construct the damage identification vector of the composite point-supported glass curtain wall.

[0027] S11. Based on principal component analysis and Euclidean distance, the damage identification vector of composite point-supported glass curtain walls is decorrelated and quantified to construct a damage degree identification index for composite point-supported glass curtain walls.

[0028] S12. The damage degree identification index of the composite point-supported glass curtain wall and the bolt preload applied to the glass panel of the composite point-supported glass curtain wall are piecewise fitted to obtain the functional relationship between the damage degree identification index of the composite point-supported glass curtain wall at the preset position and the bolt preload of the glass panel of the composite point-supported glass curtain wall.

[0029] Optionally, in the above technical solution, the process of constructing the damage identification vector of the composite point-supported glass curtain wall in S10 includes:

[0030] Construct a damage identification vector f for a composite point-supported glass curtain wall, f = [f1, f2, ..., f n ;f bar ], where there are n natural frequencies, f1, f2, ..., f n Let f represent the first natural frequency, the second natural frequency, ..., the nth natural frequency, respectively. bar The centroid frequency represents the power spectrum of the vibration signal. p (n′) f is the amplitude of the n′-th spectral line in the power spectrum of the vibration signal. n′ Let be the frequency value of the n′-th spectral line in the power spectrum of the vibration signal, and N be the total number of spectral lines in the power spectrum of the vibration signal.

[0031] Optionally, in the above technical solution, the process of constructing the damage degree identification index for the composite point-supported glass curtain wall in S11 includes:

[0032] S110. Based on the damage identification vector f calculated under various bolt preloads, construct the original damage pattern matrix A:

[0033]

[0034] Among them, f i The damage identification vector is the one collected and calculated at a preset position after applying the i-th type of bolt preload to the composite point-supported glass curtain wall, where 1≤i≤m, and m is the number of bolt preloads applied to the composite point-supported glass curtain wall. The j-th natural frequency, 1≤j≤n, is the frequency collected and calculated at a preset position after applying the i-th type of bolt preload to the composite point-supported glass curtain wall. The centroid frequency of the vibration signal power spectrum collected and calculated at a preset position after applying the i-th type of bolt pre-tightening force to the composite point-supported glass curtain wall;

[0035] S111. Standardize the original damage pattern matrix A to construct the standardized damage pattern matrix B:

[0036]

[0037] in, μ j Let σ be the mean of the j-th column of the standardized damage pattern matrix B. j Let $\mathbf{B}$ be the variance of the j-th column of the standardized damage pattern matrix $B$. μ bar σ is the mean of the (n+1)th column of the standardized damage pattern matrix B. bar Let be the variance of the (n+1)th column of the standardized damage pattern matrix B, where 1 ≤ i* ≤ m, and m* = m;

[0038] S112. Using principal component analysis, the standardized damage mode matrix B is subjected to decorrelation and dimensionality reduction operations to construct the damage mode matrix C:

[0039]

[0040] Wherein, the damage mode matrix C is an m×λ matrix, where λ depends on the cumulative contribution rate of each principal component calculated by principal component analysis, f i# This represents the new vector obtained by principal component analysis of the standardized damage mode matrix B, where 1 ≤ i# ≤ m, m# = m. The j#th damage degree identification value corresponding to the application of the i-th type of bolt preload to the composite point-supported glass curtain wall is obtained by principal component analysis of the standardized damage mode matrix B, where 1≤j#≤λ.

[0041] S113. Based on the damage pattern matrix C, construct a damage degree identification index:

[0042]

[0043] in, f is an indicator for identifying the degree of damage when the i-th type of bolt preload is applied to a composite point-supported glass curtain wall. u# This represents the row vector corresponding to the undamaged conditions in the damage mode matrix C. f u# The corresponding j#-th element.

[0044] Optionally, in the above technical solution, S12 specifically includes:

[0045] S120. Establish damage degree identification indices obtained under various bolt preloads and a polynomial fitting function f for the bolt preload applied to the glass panels of the composite point-supported glass curtain wall. ED (x)1,f ED (x)1=ax n″ +bx n″-1 +…+gx+h;

[0046] S121. Establish damage degree identification indices obtained under various bolt preloads and an exponential fitting function f for the bolt preload applied to the glass panels of the composite point-supported glass curtain wall. ED (x)2,f ED (x)2=A′+B′e C′x ;

[0047] Take f ED (x)1 and f ED The intersection point of the curve (x)2 is the dividing point, and a piecewise function f is constructed. ED(x),

[0048] Where x is the bolt preload, a, b, g, h, A′, B′ and C′ are function fitting parameters, n″ is the highest power of the polynomial fitting function, and n″≥2.

[0049] Optionally, in the above technical solution, a rubber hammer or a drone capable of firing rubber bullets is used to excite the glass panel of the composite point-supported glass curtain wall to a preset position, and a laser vibration meter is used to collect vibration signals.

[0050] Optionally, in the above technical solution, the preset position of the glass panel of the point-supported glass curtain wall to be tested is: the intersection of one-eighth of the long side and one-sixth of the short side of the glass panel of the point-supported glass curtain wall to be tested.

[0051] The preset position of the glass panel in the composite point-supported glass curtain wall is the intersection of one-eighth of the long side and one-sixth of the short side of the glass panel.

[0052] Example 2:

[0053] like Figure 2 As shown, a method for quantitatively determining the degree of damage to a point-supported glass curtain wall using a 1000mm×700mm glass panel is described. In this embodiment, a torque wrench is used to adjust the tightness of the bolts, and the specific working conditions are shown in Table 1.

[0054] Table 1:

[0055]

[0056]

[0057] In the table, ABCD represent four bolts on the glass panel: bolt A, bolt B, bolt C, and bolt D. The superscript number indicates the bolt preload, where "0" means the bolt is manually adjusted to just contact the panel, with a preload of approximately zero. The fitting function f is used for conditions 1-12. ED (x), Condition 1 is the undamaged condition, and Conditions 13 to 15 are the verification conditions.

[0058] It should be noted that in this embodiment, a laser vibrometer is used to remotely and non-contactly acquire the pulse excitation response signal of the point-supported glass curtain wall. Furthermore, to ensure a high signal-to-noise ratio, a rubber hammer is used to excite the panel when acquiring vibration signals under various bolt preload conditions.

[0059] Numerical simulations and experimental tests using Abaqus software have verified that pulse excitation and signal acquisition at the intersection of one-eighth of the long side and one-sixth of the short side of the glass panel can stably excite more modal information. For example... Figure 3 As shown, taking the six sets of data in working condition 9 of this embodiment as an example, the first six natural frequencies of the curtain wall panel can be stably obtained under this working condition. However, the subsequent frequencies are greatly affected by high-frequency noise interference and cannot be stably obtained.

[0060] It should be noted that when the bolt preload decreases, the stiffness of the glass panel decreases, leading to a decrease in the first-order natural frequency. Recent research indicates that as the bolt preload decreases, not only the first-order natural frequency decreases, but all natural frequencies also show a decreasing trend, and higher-order natural frequencies are more sensitive to damage changes than lower-order natural frequencies. Figure 4 As shown, in this embodiment, the preload of working condition 11 and working condition 12 is relatively similar, the first and second natural frequencies are completely consistent, while the third, fourth, fifth and sixth natural frequencies, which have relatively higher orders, show obvious differences.

[0061] As bolt preload decreases, indicating increased curtain wall damage, the multi-order natural frequencies decrease, which is reflected in the spectrum as a leftward shift of the peak values ​​corresponding to each natural frequency. Simultaneously, with worsening damage, the contribution of low-order modes to the curtain wall panel vibration signal increases, while the contribution of high-order modes decreases. The panel vibration will then primarily consist of the superposition of low-order modes, specifically manifested in the spectrum as a shift of spectral energy towards lower frequencies. For example... Figure 5 As shown, the characteristics of the vibration signal power spectrum change under conditions 8 and 12 in this embodiment will be explained.

[0062] The centroid frequency is a statistical index describing the change in the centroid of the power spectrum, and based on this index, the aforementioned trend can be quantitatively described. To reduce the interference of random frequency components in the power spectrum and concentrate energy near each natural frequency, this embodiment uses the Welch method to calculate the power spectrum of the vibration signal. The index pattern is as follows: Figure 6 As shown in the figure. This index takes into account the frequency variation characteristics of more orders and the energy variation trend of the entire spectrum, and therefore is significantly more sensitive to damage changes than to lower-order natural frequencies.

[0063] Because the indicators in the damage identification vector f have different sensitivities to curtain wall damage (i.e., different variability) and different data scales, if these factors are not considered, indicators with high variability or large data scales will play a decisive role, while the effects of other indicators will be masked, ultimately leading to significant deviations in the identification results. To eliminate the influence of these factors, the data is first standardized. Then, to eliminate the correlation between the indicators in the damage identification vector f, reduce information redundancy, and improve identification accuracy, principal component analysis (PCA) is used to decorrelate and reduce the dimensionality of the standardized data. Finally, Euclidean distance is used to fuse and quantify the newly generated data. Through this process, the advantages of each indicator are fully utilized, significantly improving the stability, sensitivity, and accuracy of bolt preload identification.

[0064] In this embodiment, the damage identification vector f is a 7-dimensional vector, including the 6 natural frequencies of the front panel and the centroid frequency of the power spectrum; the damage degree identification index f for each working condition relative to the undamaged working condition (i.e., working condition 1) is calculated respectively. ED The specific results are shown in Table 2.

[0065] Table 2:

[0066] Operating conditions 1 2 3 4 5 6 7 8 <![CDATA[f ED ]]> 0.00 0.06 0.29 0.10 0.94 1.38 2.26 2.70 Operating conditions 9 10 11 12 13 14 15 \ <![CDATA[f ED ]]> 3.77 5.38 7.32 7.99 0.04 1.70 3.34 \

[0067] Under different bolt tightening conditions, the sensitivity of each natural frequency and centroidal frequency to changes in preload varies considerably, thus leading to f ED The relationship curves between bolt preload F and the preload force F under different preload conditions are different. When the total bolt preload force is small, f... ED The relationship between the bolt preload and the preload is approximately a polynomial function; however, when the bolt preload increases to a certain extent, f... ED The sensitivity to changes in preload decreases, and the curve relationship approximates an exponential function.

[0068] Depend on Figure 7 As shown, using the data obtained from operating conditions 1 to 12, the index f proposed in this application is... ED A comparison was made between the first-order natural frequency, a conventional indicator for identifying the degree of curtain wall damage, and the bolt preload identification accuracy. As shown in the figure, when the bolt preload decreases from 64 N·m to 0 N·m, the first-order natural frequency changes from 25.20 Hz to 20.8 Hz, a change of only 17.44%, while the indicator f proposed in this application... ED The value changes from 0.00 to 7.99, a very large relative rate of change, significantly improving the ability to identify changes in bolt preload. Furthermore, relative to the first-order natural frequency, f... ED Even when the bolt preload is high, it still has a good ability to detect changes in preload.

[0069] like Figure 8As shown, based on the data analysis of working conditions 1 to 12 in this embodiment, the function f is determined according to the fitting results. ED The equation for (x) is:

[0070]

[0071] In this embodiment, the f corresponding to working conditions 13-15 is... ED Input function f ED In (x), the calculated bolt preload x is obtained, and the identification accuracy is shown in Table 3;

[0072] Table 3:

[0073] Operating conditions 13 14 15 Actual value (N·m) 61.00 42.00 33.00 Estimated value (N·m) 60.68 40.94 33.56 Relative error / % 0.52 2.52 1.70

[0074] Analysis of curve fitting effect and recognition error shows that the method proposed in this application has a good curve fitting effect and high recognition accuracy, which meets the actual needs of engineering.

[0075] In the above embodiments, although the steps are numbered S1, S2, etc., they are only specific embodiments given in this application. Those skilled in the art can adjust the execution order of S1, S2, etc. according to the actual situation, which is also within the protection scope of this invention. It can be understood that in some embodiments, some or all of the above embodiments may be included.

[0076] Those skilled in the art will know that this invention can be implemented as a system, method, or computer program product.

[0077] Therefore, this disclosure can be implemented in the following forms: it can be entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, generally referred to herein as a "circuit," "module," or "system." Furthermore, in some embodiments, the invention can also be implemented as a computer program product in one or more computer-readable media, the computer-readable medium containing computer-readable program code.

[0078] Any combination of one or more computer-readable media may be used. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device.

[0079] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for quantitatively determining the damage degree of a point-supported glass curtain wall, characterized in that, The method comprises the following steps: obtaining a function relationship between a damage degree identification index of a preset position of a glass panel of the replica point-supported glass curtain wall and a bolt pre-tightening force of the glass panel of the replica point-supported glass curtain wall by using a replica point-supported glass curtain wall which is completely the same as the to-be-tested point-supported glass curtain wall; calculating the damage degree identification index of the preset position of the glass panel of the to-be-tested point-supported glass curtain wall, and calculating the current bolt pre-tightening force of the glass panel of the to-be-tested point-supported glass curtain wall by using the function relationship, wherein the position of the preset position of the glass panel of the to-be-tested point-supported glass curtain wall is the same as the position of the preset position of the glass panel of the replica point-supported glass curtain wall; the process of obtaining the function relationship comprises the following steps: collecting vibration data of the preset position of the glass panel of the replica point-supported glass curtain wall under different bolt pre-tightening forces, calculating multi-order natural frequencies and gravity frequencies of vibration signal power spectrums, and constructing a damage identification vector of the replica point-supported glass curtain wall; performing de-correlation and quantization on the damage identification vector of the replica point-supported glass curtain wall based on a principal component analysis method and a Euclidean distance, and constructing a damage degree identification index of the replica point-supported glass curtain wall; performing piecewise fitting on the damage degree identification index of the replica point-supported glass curtain wall and the bolt pre-tightening force applied to the glass panel of the replica point-supported glass curtain wall, and obtaining the function relationship between the damage degree identification index of the preset position of the replica point-supported glass curtain wall and the bolt pre-tightening force of the glass panel of the replica point-supported glass curtain wall.

2. The method according to claim 1, wherein The process of constructing the damage identification vector of the replica point-supported glass curtain wall comprises the following steps: Constructing a damage identification vector for composite point-supported glass curtain walls , There are n natural frequencies in total. , , ..., These represent the first natural frequency, the second natural frequency, and so on. n First natural frequency, This represents the centroid frequency of the power spectrum of the vibration signal. , The first in the power spectrum of the vibration signal The amplitude of each spectral line, The power spectrum of the vibration signal is the first Frequency values ​​of the spectral lines N This represents the total number of spectral lines in the power spectrum of the vibration signal.

3. The method according to claim 2, wherein the method is characterized by, The process of constructing the damage degree identification index of the replica point-supported glass curtain wall comprises the following steps: According to the damage identification vectors calculated under a plurality of different bolt pre-tightening forces , a raw damage mode matrix A is constructed: wherein, a damage identification vector collected and calculated at the preset position after a first i number of bolt pre-tightening forces is applied to the duplicate point-supported glass curtain wall, 1≤i≤m, m a number of bolt pre-tightening forces applied to the duplicate point-supported glass curtain wall; a damage identification vector collected and calculated at the preset position after a first i number of bolt pre-tightening forces is applied to the duplicate point-supported glass curtain wall, j a first-order natural frequency collected and calculated at the preset position after a first 1≤j≤n , a first-order natural frequency collected and calculated at the preset position after a first i number of bolt pre-tightening forces is applied to the duplicate point-supported glass curtain wall; standardizing the original damage mode matrix A to construct a standardized damage mode matrix B: wherein, , is the mean of the n+1th column of the normalized damage mode matrix B, j is the mean of the n+1th column of the normalized damage mode matrix B, is the variance of the n+1th column of the normalized damage mode matrix B, j is the variance of the n+1th column of the normalized damage mode matrix B, , is the mean of the n+1th column of the normalized damage mode matrix B, is the mean of the n+1th column of the normalized damage mode matrix B, 1≤ ≤m， =m； performing de-correlation and dimension reduction operations on the standardized damage mode matrix B by using a principal component analysis method to construct a damage mode matrix C: The damage mode matrix C is a matrix, depending on the cumulative contribution rate of each principal component calculated by principal component analysis, a new vector obtained by principal component analysis of the normalized damage mode matrix B, 1≤ ≤m， =m， an i-th damage degree identification value corresponding to the i-th bolt pre-tightening force applied to the duplicate point-supported glass curtain wall, wherein i is an integer between 1 and n, i 1≤ ≤ ； ​​ constructing a damage degree identification index according to the damage mode matrix C: wherein, A damage degree identification index corresponding to the damage mode matrix C after a bolt pre-tightening force is applied to the double-point-supported glass curtain wall, i A damage degree identification index corresponding to the damage mode matrix C after a bolt pre-tightening force is applied to the double-point-supported glass curtain wall, A row vector corresponding to an undamaged working condition in the damage mode matrix C, Indicates The first Element corresponding.

4. The method according to claim 3, wherein the method is characterized by, performing piecewise fitting on the damage degree identification index of the replica point-supported glass curtain wall and the bolt pre-tightening force applied to the glass panel of the replica point-supported glass curtain wall, and obtaining the function relationship between the damage degree identification index of the preset position of the replica point-supported glass curtain wall and the bolt pre-tightening force of the glass panel of the replica point-supported glass curtain wall, comprising the following steps: A polynomial fitting function of a plurality of different bolt pre-tightening forces and a plurality of damage degree recognition indexes obtained under the plurality of different bolt pre-tightening forces is established , ; Establishing an exponential fitting function of the damage degree recognition index obtained under a plurality of different bolt pre-tightening forces and the bolt pre-tightening force applied to the glass panel of the complex point-supported glass curtain wall , ; Take and the intersection of the curve as a demarcation point, construct a piecewise function , ; wherein is the bolt pre-tightening force, a, b, g, h, 、 and is a function fitting parameter, is the highest power of the polynomial fitting function, ≥2 .

5. The method according to any one of claims 1 to 4, wherein the point-supported glass curtain wall damage degree is quantitatively determined. vibrating the preset position of the glass panel of the replica point-supported glass curtain wall by using a rubber hammer or a drone capable of launching rubber bullets, and collecting vibration signals by using a laser vibration measuring instrument.

6. The method according to any one of claims 1 to 4, wherein the point-supported glass curtain wall damage degree is quantitatively determined. The preset position of the glass panel of the to-be-tested point-supported glass curtain wall is an intersection position of one-eighth of a long side and one-sixth of a short side of the glass panel of the to-be-tested point-supported glass curtain wall. The preset position of the glass panel of the replica point-supported glass curtain wall is an intersection position of one-eighth of a long side and one-sixth of a short side of the glass panel of the replica point-supported glass curtain wall.

Images