Method for positioning lowest-strength area of wood based on ultrasonic detection and visual inspection parameters

Abstract

The invention discloses a wood lowest strength area positioning method based on ultrasonic detection and visual inspection parameters, and belongs to the technical field of wood structure engineering. The positioning method comprises the following steps: calculating the propagation velocity v of ultrasonic waves in wood; the dynamic elastic modulus E of the wood is calculated, and rho is the air-dry density of the wood; measuring the size K of the maximum wood section in a flat bending three-point bending test span l, and meanwhile, placing the maximum wood section near a loading point; carrying out flat bending three-point bending damage to obtain the three-point bending strength f of each test material; and substituting actual measurement samples fi, Ei and Ki into the multiple linear regression model to estimate a regression coefficient of the multiple linear regression model, positioning the lowest-strength areas of other woods according to an obtained multiple regression equation, and determining the positions of four-point (three-point) bending test loading points. The positioning method provided by the invention is universal and feasible, and can realize accurate positioning of the lowest strength area before a four-point (three-point) bending test.

Description

Technical Field

[0001] The present invention relates to a method for locating the lowest strength region of wood based on ultrasonic detection and visual parameters, which is applicable to the location detection of the region between or near the loading points in the four-point (or three-point) bending test during the derivation and verification of the mechanical grading settings of structural timber or the test of the flexural performance of structural timber. It belongs to the technical field of wood structure engineering. Background Art

[0002] As a natural biomass material, wood is widely used in fields such as construction, furniture, and bridges due to its renewable nature, light weight and high strength, and environmental protection characteristics. However, there are natural defects (such as knots, cracks, decay) and microstructural heterogeneities (such as density gradients, fiber orientation deviations) inside wood, which will lead to uneven distribution of its mechanical properties and significant reduction of local strength. As structural timber, it is necessary to grade its strength. One grading method is visual grading, which mainly grades through various defect indicators on the wood surface; another grading method is mechanical grading, that is, through some non-destructive testing methods, such as ultrasonic method, stress wave method, vibration method, bending stress method, etc. to measure the dynamic elastic modulus of wood. Both grading methods need to establish a regression model between visual defects or dynamic elastic modulus and flexural elastic modulus and flexural strength through the static four-point (or three-point) bending test of wood. However, in the bending test, whether it is the national standard or foreign standards (such as, GB / T 26899, ASTM D143 standard), it is required to place the maximum downgrading defect between the loading points. This process is all about manually observing the color and texture changes on the wood surface or visible defects (such as knots, wormholes), and combining experience to judge the strength weak area, so as to locate the position of the maximum downgrading defect. However, this method has significant deficiencies. The key point is that hidden defects such as internal cracks and decay in wood cannot be discovered by visual inspection, and tiny defects (such as micro-cracks) not discovered inside the wood may form more dangerous stress concentration points. Some studies have shown that such defects may make the actual strength 50% - 70% lower than the visually predicted value, resulting in a large deviation between the loading position and the actual weakest position in the bending test. This causes problems in deriving and establishing the wood strength grading model and settings, such as overestimating the wood strength, which is likely to result in inaccurate strength grading and there are certain potential safety hazards. Therefore, how to provide a method for the lowest strength region that can be more accurate than the existing visual method is an urgent problem to be solved. Summary of the Invention

[0003] The technical problem to be solved by the present invention is that the present invention provides a method for locating the lowest strength region of wood based on ultrasonic detection and visual parameters, which is general and feasible to accurately locate the lowest strength region before the four-point (three-point) bending test.

[0004] To solve the above technical problem, the technical solution adopted by the present invention is:

[0005] The present invention uses professional non - metallic ultrasonic testing instruments (such as the MC - 6310 non - metallic ultrasonic detector produced by Beijing Mingchuang Technology Co., Ltd., and the Sylvatest4 ultrasonic wood detector) combined with visual parameters to more accurately locate the area with the lowest strength.

[0006] The non - metallic ultrasonic testing instrument has an adjustable emission voltage (not less than 50V). The adjustable emission voltage can control the distance between the detection transducers according to the specific detection area. And, a broadband transducer is used, and the bandwidth should at least cover the range of 20 - 200 kHz to adapt to the absorption and scattering characteristics of different tree species and wood properties for ultrasonic waves of different frequencies, improving the accuracy and reliability of detection.

[0007] A method for locating the lowest strength area of wood based on ultrasonic detection and visual parameters includes the following steps:

[0008] The single - sided flat - measurement method can be used in the test location process. That is, the ultrasonic transmitting transducer and the receiving transducer are placed on the same surface of the wood, maintaining a certain distance. By moving the transducers, the ultrasonic travel time (τ, unit: ms) at different positions is measured. According to the distance (s, unit: mm) between the transmitting transducer and the receiving transducer, the ultrasonic propagation speed (v, unit: m / s) is calculated according to formula (1).

[0009] v = s / τ (1)

[0010] The relationship between the dynamic elastic modulus E (unit: MPa) of wood and the sound speed v (unit: m / s) is shown in formula (2). Where ρ is the air - dried density of wood, unit: kg / m 3 。

[0011] E = ρv 2 ·10 -6 (2)

[0012] Then, for the same batch of test specimens with the same nominal size, a certain number of test specimens (not less than 50) are randomly selected for a three - point bending test of flat bending. Before the bending test, the maximum knot size within the span should be measured. At the same time, the largest knot is placed near the loading point, and the transducers are placed at the support point position of the test specimen to measure the ultrasonic propagation speed in this area, and then the dynamic elastic modulus E is calculated through formula (2). Where the knot size is defined as the ratio of the knot on a single surface of the wood to the width of this surface, and the maximum value including the four surfaces of the wood is the maximum knot size K.

[0013] After that, three-point bending failure under flat bending is carried out on each test specimen. Among them, the span (l, unit: mm) is at least 18 times the height of the wood (h, unit: mm). The wood height h is the wood dimension parallel to its loading direction. In the flat bending test, the wood height h is the wood thickness t; b is the wood width, unit: mm; P max is the maximum load, unit: N. Calculate according to formula (3) to obtain the three-point bending strength of each test specimen.

[0014]

[0015] Thus, a multiple linear regression model is obtained with the dynamic elastic modulus (E) and the maximum knot size (K) as explanatory variables and the wood bending strength (f) as the target variable. According to the measured samples f i , E i , K i to estimate the regression coefficients of the multiple regression model, f i , E i , K i are the bending strength, dynamic elastic modulus, and maximum knot size of the i-th measured sample respectively, as shown in formula (4).

[0016] f i =β0 + β1·E i + β2·K i + ε i , i = 1, 2, …, n (4)

[0017] where n is the number of samples, β0, β1, β2 are the regression coefficients to be estimated, and ε i is the random error term between the predicted bending strength and the actual bending strength, and it is assumed that ε i are independent of each other and follow a normal distribution with a mean of 0 and a variance of σ 2 , that is, ε i ~N(0, σ 2 ).

[0018] N is the abbreviation of Normal Distribution, representing the normal distribution.

[0019] It can be seen from this that f i follows a normal distribution, f i ~N(β0 + β1·E i + β2·K i , σ 2 ); Since f i are independent of each other, their joint probability density function is the product of their respective probability density functions:

[0020]

[0021] Its likelihood function L is:

[0022]

[0023] Taking the logarithm of the likelihood function L, we get the log-likelihood function lnL:

[0024]

[0025] For β0, β1, β2, σ 2 Find the partial derivative and make it equal to 0. The system of simultaneous equations can be written in matrix form X T Xβ=X T Y, where X is the design matrix, β = [β0, β1, β2] T , Y=[f1,f2,…,f n ] T By solving this set of equations, we can obtain the maximum likelihood estimates of β0, β1, and β2.

[0026] The design matrix X is:

[0027]

[0028] Among them, E1, E2, …, E n Represent the dynamic elastic modulus of n samples, K1, K2, ..., K n Represent the maximum knot sizes of n samples respectively.

[0029] but, in,

[0030] The final result will be Substituting into the regression model formula (4), the multivariate linear regression equation for predicting the bending strength of wood using the dynamic elastic modulus and the maximum knot size measured by ultrasonic wave is obtained:

[0031]

[0032] In specific applications, a regression model combining more visual parameters (such as texture angle, cracks, and annual ring width) can also be established by adding corresponding explanatory variables to the above regression model (Formula 4). The method for solving the estimated regression coefficients can also be carried out using the above theoretical derivation.

[0033] After that, the lowest strength area in the same batch of wood is predicted according to the obtained regression equation. Taking the national standard "GB / T 36407 Mechanically Stress-Graded Lumber" as an example, during the measurement of the modulus of elasticity in bending and the bending strength of the narrow face of the sawn timber, in the four-point bending test, the span (l) is at least 18 times the height (h) of the wood. The height of the wood is the dimension of the wood parallel to the loading direction. Here, the height of the wood is equal to the width (b). Additionally, the distance from the two end support points to the ends of the wood is h / 2. The test adopts a three-point loading method, that is, the distance a between the loading points is l / 3. Taking one end of the wood as the origin of the one-dimensional coordinate and the length direction of the wood as the x-axis, the area that may contain the loading points is Figure 1 Region 1 in Figure 1 , and its interval D is from 6.5h to L - 6.5h. L is the length of the wood.

[0034] As Figure 2 shown, the wood is bisected into two test areas. The first area is marked as A1, and the second area is A2. Additionally, a third area (A3) is added such that A3 has the same length as the previous two areas and its midpoint position coincides with the midpoint position of the wood. In the three areas, the dynamic modulus of elasticity and the maximum knot size are measured, and the predicted bending strength is calculated using formula (9). The lowest strength area among the three areas is compared and determined;

[0035] According to the above test area division method, the previously determined lowest strength area is divided into three different areas again and the lowest strength area is determined again.

[0036] According to the above steps, until the length of the lowest strength area is less than or equal to the distance between the specified loading points in the four-point bending test, and the left endpoint coordinate of the interval of this area should not be less than 6.5h and the right endpoint coordinate should not be greater than L - 6.5h (it should not intersect with Figure 1 Region 2 specified in Figure 1 ), then this area is the area where the lowest strength position of the wood is located; if it is a three-point bending test, the interval range of the area where the lowest strength position is located should not be higher than 2h or as small as possible depending on the specific situation.

[0037] In the present invention, a data set for predicting the bending strength from knots and the dynamic modulus of elasticity is obtained through a three-point bending test. The three-point bending test is a flat bending test, that is, the wide face of the wood is facing up for bending. In this way, the loading direction of the testing machine is parallel to the thickness direction of the wood, and at this time the thickness of the wood is the height of the wood; for the four-point bending test, the sawn timber generally needs to be bent sideways, that is, the narrow face is facing up for loading. In this case, the height direction of the wood is the width of the wood. The position of the loading point is the position that needs to be predicted in the present invention. In actual tests, the lowest strength area needs to be placed between the loading points.

[0038] The present invention has the following beneficial effects:

[0039] The method for locating the lowest strength area of ​​wood based on ultrasonic detection and visual parameters of the present invention comprehensively considers the factors of wood knots and ultrasonic waves, has the highest determination coefficient, and the prediction results are more real and reliable, with fewer false positive prediction errors. The loading position in the bending test is close to the actual weakest position, which is conducive to the accurate grading of wood strength and accurate evaluation of mechanical properties. It is more effective than the existing method and avoids some safety hazards caused by overestimation of its strength due to incorrect determination of the lowest strength position of wood. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is a schematic diagram of the area where the loading point of the bending test wood of the present invention is located;

[0041] Figure 2 This is a schematic diagram of the present invention gradually approaching the lowest intensity region. DETAILED DESCRIPTION

[0042] The following describes the embodiments of the present invention through specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention.

[0043] Example 1

[0044] A method for locating the lowest strength area of ​​wood based on ultrasonic detection and visual parameters.

[0045] Taking domestic fir sawn timber as an example, 50 sawn timbers were randomly selected with the nominal dimensions of: length L = 3000 mm, width b = 90 mm, thickness t = 35 mm.

[0046] The above sawn timbers are numbered and weighed in turn, and the density (ρ) of each test piece, i.e., the air-dry density of the wood, is calculated. The density of the wood is determined by dividing the mass (m) of each test piece by its volume.

[0047] Each sawn timber was divided into four areas, each area was 750 mm long, and then cut into four test materials, each of which was marked with serial numbers 1 to 4.

[0048] The maximum knot size (K) of each test material was measured with a ruler, and the dynamic elastic modulus was calculated using a non-metallic ultrasonic instrument (simultaneous calculation of formula (1) and formula (2)).

[0049] After that, three-point bending failure under flat bending was carried out on each test specimen, and its flexural strength was measured (Formula (3)). One test specimen was randomly selected from each sawn timber, and a total of 50 test specimens were used as the data set for constructing a multiple regression model, so as to construct a multiple regression model for predicting flexural strength with dynamic elastic modulus and maximum knot size.

[0050] According to Formulas (4) to (8), the calculated results are 28.0612, -0.00324, and -8.5714 respectively, that is, the regression equation is Coefficient of determination R 2 = 0.5016. After that, the above equation was used to determine the lowest strength area of the remaining sawn timbers in the same batch of wood.

[0051] Taking the measurement process of the flexural elastic modulus and flexural strength of the narrow surface of sawn timber specified in the national standard "GB / T 36407 Machine Stress-Graded Sawn Timber" as an example, the four-point bending span (l) is at least 18 times the height (h) of the wood. The distance from the support point to the end of the wood is h / 2, and the height of the wood is the wood dimension parallel to its loading direction, that is, h = b = 90 mm. In this embodiment, l = 1620 mm is taken. The three-point loading method is adopted, that is, the distance a between the loading points = l / 3 = 540 mm.

[0052] As Figure 2 shown, the positioning process of one of the test specimens is described. The maximum knot size and dynamic elastic modulus of the test area A1 are 0.33 and 9770 MPa respectively, the maximum knot size and dynamic elastic modulus of A2 are 0.14 and 8477 MPa respectively, and the maximum knot size and dynamic elastic modulus of A3 are 0.25 and 7405 MPa respectively. According to the regression equation (that is, Formula (9)), the strengths of A1, A2, and A3 are predicted to be 36.44, 38.77, and 29.83 MPa respectively. Therefore, the predicted strength of the A3 area is the lowest. Since the length of A3 is greater than a, it is necessary to reduce the length of the predicted lowest strength area;

[0053] Then, A3 was further divided into B1, B2, and B3. Among them, the lengths of B1, B2, and B3 are all 750 mm. Similarly, according to the above steps, the maximum knot size and dynamic elastic modulus of each area were measured, and then according to the regression equation (that is, Formula (9)), the strengths of B1, B2, and B3 were predicted to be 28.76, 27.48, and 29.10 MPa respectively. Therefore, the predicted strength of the B2 area is the lowest. Since the length of B2 is greater than a, the range of the predicted maximum downgrading area is further reduced;

[0054] Further, divide B2 into C1, C2, and C3. Among them, the lengths of C1, C2, and C3 are all 375 mm. Similarly, according to the above steps, measure the maximum knot size and dynamic elastic modulus of each region, and then predict the strengths of C1, C2, and C3 to be 28.28, 26.53, and 26.24 MPa respectively according to the regression equation (i.e., formula (9)). Therefore, the predicted strength of the C3 region is the lowest. Since the length of the C3 region is less than a, and check that the C3 region does not intersect with Figure 2 the area 2 shown.

[0055] Finally, the C3 region is the determined lowest-strength region.

[0056] Comparative example

[0057] To compare the advantages of the method of the present invention in determining the lowest-strength region, the data set in this comparative example is processed as follows:

[0058] Take the lowest flexural strength measured by the three-point bending failure test of each test specimen (a total of 4) cut from each sawn timber in the embodiment as the actual lowest flexural strength.

[0059] After that, using the dynamic elastic modulus, maximum knot size, and flexural strength of each measured test specimen, establish regression equations for predicting the strength of wood by separately using the maximum knot size to predict the strength of wood and separately using the dynamic elastic modulus measured by ultrasonic waves to predict the strength of wood, that is, formula (4) is transformed into a simple unary linear regression model, as shown in formulas (10) and (11) respectively.

[0060] f i =α0 + α1·E i +ε E，i , i = 1, 2, …, n (10)

[0061] f i =γ0 + γ1·K i +ε K，i , i = 1, 2, …, n (11)

[0062] Among them, formula (10) is the regression equation established for predicting the strength of wood by separately using the dynamic elastic modulus measured by ultrasonic waves; formula (11) is the regression equation established for predicting the strength of wood by separately using the maximum knot size (a visual parameter); a0, a1, γ0, γ1 are the regression coefficients to be estimated in the two unary linear regression models respectively, and ε E,i , ε K,i are the error terms between the actual flexural strength and the predicted flexural strength corresponding to the two regression models respectively.

[0063] According to the similar theoretical calculation process of formulas (4) to (8), the regression equation for predicting the strength of wood by separately using the dynamic elastic modulus can be obtained as: R 2 = 0.3054, and the regression equation for predicting the strength of wood using only the maximum knot size is R 2 = 0.1903.

[0064] The measured dynamic elastic modulus or the maximum knot size of each test specimen cut from 50 pieces of sawn timber was respectively substituted into the above regression equations, and the measured dynamic elastic modulus and the maximum knot size were simultaneously substituted into the multiple linear regression equation obtained in Example 1 to predict the bending strength of 4 test specimens on the same piece of sawn timber. If the test specimen with the lowest predicted strength is the same as the test specimen with the actual lowest strength, it proves that the prediction is correct. On the contrary, if it is predicted that a certain test specimen on the same piece of sawn timber has the lowest strength, but the actual test specimen with the lowest strength does not match it, it proves that the prediction is incorrect (false positive). The specific results are shown in Table 1 below.

[0065] Table 1 Coefficient of determination and number of false positives for predicting the lowest bending strength of wood by different methods

[0066] Evaluation index Knot Ultrasonic wave Knot + Ultrasonic wave <![CDATA[Coefficient of determination (R 2 )]]> 0.1903 0.3054 0.5016 Number of false positives (roots) 24 18 9

[0067] It should be understood that, in order to streamline the present disclosure and assist in understanding one or more of the various inventive aspects, in the above description of the exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together into a single embodiment or the description thereof. However, the disclosed method should not be construed as reflecting the intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as reflected in the claims, the inventive aspects lie in less than all of the features of the previously disclosed embodiments. Thus, the claims following the detailed description are hereby expressly incorporated into the detailed description, where each claim stands on its own as a separate embodiment of the present invention.

[0068] Although the present invention has been described in terms of a limited number of embodiments, those skilled in the art in this technical field will appreciate, from the above description, that other embodiments can be contemplated within the scope of the present invention thus described. In addition, it should be noted that the language used in this specification has been principally selected for readability and teaching purposes rather than for the purpose of interpreting or limiting the subject matter of the present invention. Thus, many modifications and variations will be apparent to those of ordinary skill in the art in this technical field without departing from the scope and spirit of the appended claims. For the scope of the present invention, the disclosure of the present invention is illustrative, not restrictive, and the scope of the present invention is defined by the appended claims.

[0069] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A method for locating the lowest strength area of ​​wood based on ultrasonic detection and visual parameters, characterized in that: The following steps are involved: Step 1, calculate the propagation speed v of ultrasonic waves in wood; Step 2, calculate the dynamic elastic modulus E of the wood, where ρ is the air-dry density of the wood; E=ρv 2 ·10 -6 (2); Step 3, measuring the span of the flat three-point bending test and the maximum wood knot size K, and placing the largest wood knot near the loading point; Step 4: Perform three-point bending failure and calculate the three-point bending strength f of each specimen according to formula (3): Where l is the span, h is the height of the wood, b is the width of the wood, P max is the maximum load of the bending test; Step 5: According to the measured sample f i , E i , K i Substitute into the multiple linear regression model to estimate the regression coefficient of the multiple linear regression model, f i , E i , K i are the measured flexural strength, dynamic elastic modulus, and maximum knot size of the i-th sample respectively; f i =β0+β1·E i +β2·K i +e i ,i=1,2,…,n (4) Where n is the sample size, β0, β1, β2 are the regression coefficients to be estimated, and ε i is the random error term between the predicted flexural strength and the actual flexural strength, and assuming that ε i Independent of each other and subject to mean 0 and variance σ 2 The normal distribution of i ~N(0,σ 2 ); It can be seen from this that f i It follows a normal distribution, f i ~N(β0+β1·E i +β2·K i , σ 2 ); due to f i Independent of each other, their joint probability density function is the product of their respective probability density functions: Its likelihood function L is: Taking the logarithm of the likelihood function L, we get the log-likelihood function lnL: For β0, β1, β2, σ 2 Find the partial derivative and make it equal to 0. The system of simultaneous equations can be written in matrix form X T Xβ=X T Y, where X is the design matrix, β = [β0, β1, β2] T , Y=[f1,f2,…,f n ] T By solving this set of equations, we can obtain the maximum likelihood estimates of β0, β1, and β2. The design matrix X is: Among them, E1, E2, …, E n Represent the dynamic elastic modulus of n samples, K1, K2, ..., K n Respectively represent the maximum knot size of n samples; but, in, The final result will be Substituting into the multiple linear regression model formula (4), the multiple linear regression equation for predicting the bending strength of wood using the dynamic elastic modulus and the maximum knot size measured by ultrasonic wave is obtained:

2. The method according to claim 1, characterized in that In step 1, the test positioning process adopts the single-sided flat measurement method, that is, the ultrasonic transmitting transducer and the receiving transducer are placed on the same surface of the wood, maintaining a certain distance, and by moving the transducer, the ultrasonic sound time τ at different positions is measured. According to the distance s between the transmitting transducer and the receiving transducer, the ultrasonic propagation speed v is calculated according to formula (1): v=s / τ (1).

3. The method according to claim 2, characterized in that Use non-metallic ultrasonic testing equipment for ultrasonic testing.

4. The method according to claim 3, characterized in that: The non-metallic ultrasonic testing instrument has an adjustable transmitting voltage of not less than 50V and adopts a wide-band transducer with a bandwidth of at least 20 to 200kHz.

5. The method according to claim 1, characterized in that Visual inspection parameters include maximum knot size K, grain angle, cracks, and annual ring width.

6. The method according to claim 1, characterized in that In step 4, the span l is at least 18 times the height h of the wood, where the height h of the wood is the size of the wood parallel to its loading direction. In the flat bending test, the height h of the wood is the thickness t of the wood.

7. The method according to claim 1, characterized in that In step 5, the multiple linear regression model uses elastic modulus E and maximum knot size K as explanatory variables and wood bending strength f as the target variable.

8. Use of the method according to any one of claims 1 to 7 in locating the lowest strength area of ​​wood.

9. The use according to claim 8, characterized in that: Divide the wood into two test areas, the first area is marked as A1, the second area is marked as A2, and a third area A3 is added, so that the length of A3 is equal to that of the first two areas, and the midpoint position of A3 coincides with the midpoint position of the wood; in the three areas, measure the dynamic elastic modulus and the maximum wood knot size, calculate the predicted bending strength using formula (9), and compare and determine the lowest strength area among the three areas; According to the above test area division method, the previously determined lowest intensity area is divided into three different areas again and the lowest intensity area is determined again; Follow the above steps until the length of the lowest strength area is less than or equal to the distance between the loading points specified in the four-point bending test, and the coordinates of the left endpoint of the area are not less than 6.5h, and the coordinates of the right endpoint are not greater than L-6.5h. This area is the area where the lowest strength position of the wood is located. If it is a three-point bending test, the range of the area where the lowest strength position is located shall not exceed 2h or be as small as possible depending on the specific situation.

Images