A method for measuring the age of ancient building wood based on distributed activation energy model

Abstract

The present invention provides a method for measuring the age of ancient building wood based on a distributed activation energy model. By sampling at the milligram level and carrying out thermogravimetric tests at multiple heating rates, the DAEM kinetic parameters of the sample to be tested containing three main components of lignin, cellulose and hemicellulose are solved based on the distributed activation energy model and the developed DAEM multi-parameter optimization program, and then the corresponding aging time is obtained according to the established kinetic parameters and the database matching of time aging. The present invention can realize the rapid and accurate measurement of wood age in various scenarios, has the advantages of convenient operation and easy promotion, can be widely used, and can provide strong evidence for the historical age and cultural heritage of cultural relics.

Description

Technical Field

[0001] The invention relates to the field of wood age measurement, and in particular to a method for measuring the age of ancient building wood based on a distributed activation energy model. Background Art

[0002] Wood has the advantages of being beautiful and durable, having excellent processing performance, being cheap and renewable, and is widely used in the fields of building materials, furniture, and handicrafts. Especially in ancient buildings, wood plays an important role in both the load-bearing system and the functional decoration system. How to quickly and accurately obtain the age of wood and determine the service life of wood is not only of great value in the field of ancient architecture, but also an important issue in the fields of archaeology, cultural relics, and biomass.

[0003] Wood is a lignified tissue formed by plants that can grow secondary. It is an organic composite material composed of natural polymer compounds such as lignin, cellulose, and hemicellulose. In addition, it also contains secondary components such as wood extracts, including resins, gums, essential oils, pigments, alkaloids, fats, waxes, sugars, starches, and silicates. In the widely existing ancient buildings, old wood is used in large quantities, but it is susceptible to erosion by bacteria, insects, ultraviolet radiation, water, corrosive media environments, antiseptic materials, etc., and changes in properties such as density, moisture content, mechanical properties, cell morphology, physical pore structure, chemical component content, molecular structure, and molecular weight distribution, which leads to discoloration, corrosion, and functional degradation of old wood. Among various aging erosions, time determines the degree of aging of wood. The accurate determination of aging time helps us to make judgments on the cultural and historical value of ancient buildings. However, in practical applications, limited by existing measurement technology, it is difficult to obtain the aging time of wood under the combined influence of uncertainty in material history, unknown original mechanical properties, and other aging effects.

[0004] As we all know, the cambium of tree trunks grows every year, forming concentric circles of varying density on the cross-section of the trunk. Most temperate tree species form one growth ring a year, so the age of the tree can be determined based on the number of growth rings. This method is also called tree-ring dating. However, this rule is only applicable to determining the age of ancient trees and wood with intact bark. Ancient trees that are widely used in ancient buildings, wooden crafts, etc. are usually cut and processed from logs, making it difficult to determine their relative age based on complete growth ring information. The carbon 14 dating method can identify the age of wood by measuring the degree of carbon 14 decay in tree samples. However, this method requires the selection of a large number of suitable samples, takes a long time to measure, has cumbersome steps, and the special measuring device used is not easy to obtain. Summary of the invention

[0005] In view of this, the present invention proposes a method for measuring the age of ancient building wood based on a distributed activation energy model. By extracting a trace amount (milligram level) of the wood sample to be studied, thermogravimetric analysis tests are carried out at multiple heating rates, and the multi-parameter optimization algorithm and the distributed activation energy model are used to decouple the pyrolysis reaction data, the kinetic parameters of the three pseudo-components in the sample are obtained, and then the age of the wood sample is judged according to the obtained kinetic parameters of each component. The method can realize the rapid measurement of the age of the wood, thereby making up for the shortcomings of the existing wood age measurement technology such as large sampling, long time, complex operation and high cost, and has great practical application value.

[0006] To achieve the above object, the technical solution of the present invention is achieved as follows:

[0007] A method for measuring the age of ancient building wood based on a distributed activation energy model, the method comprising the following steps:

[0008] (1) Sample preparation:

[0009] Select the wood samples of known years obtained from the ancient building complex, grind them into powder, sieve them, dry them and weigh them in equal parts for later use;

[0010] (2) Thermogravimetric test:

[0011] Carry out thermogravimetric tests on the samples at different heating rates, draw TG and DTG curves at different heating rates, and analyze the reliability of the test data;

[0012] (3) DAEM parameter calculation:

[0013] Based on the thermogravimetric results at multiple heating rates, a distributed activation energy prediction model for n-order reactions containing three pseudo-components was established. The developed DAEM multi-parameter optimization program was used to solve the problem, obtain the optimal DAEM parameter group for the sample to be tested, and determine the kinetic parameters of each pseudo-component.

[0014] (4) Establish database:

[0015] Repeat the above steps (1)-(3) to calculate the kinetic parameters of each component of wood of the same species but different years, so as to establish a database of the corresponding relationship between time and kinetic parameters of wood of known years;

[0016] (5) Determine the age of the wood:

[0017] Select a part of the wood to be tested that is not contaminated by corrosion and does not cause damage to the sample, remove dust and impurities on the surface of the sample, and use a tool to cut or drill the sample to be tested; then repeat the above steps (1)-(3) to calculate the kinetic parameters of each component, and determine the age of the wood to be tested by comparing it with the established database.

[0018] Furthermore, the three pseudo components refer to lignin, cellulose and hemicellulose, respectively; and the kinetic parameters refer to the activation energy, pre-exponential factor and weight factor of lignin, cellulose and hemicellulose.

[0019] Furthermore, after ball milling in step (1), the product is sieved through a 40-60 mesh sieve, dried in an oven at 75-105° C. for 24-48 hours, and taken out. Each sample is weighed 3-5 mg for later use.

[0020] Furthermore, in step (2), 3-4 heating rates are selected, and the heating range is set to 25-850°C.

[0021] Furthermore, in step (3), the developed DAEM multi-parameter optimization program is based on the coupling of particle swarm optimization and genetic algorithm, and simultaneously solves the α and dα / dT data under multiple groups of heating rates to avoid the pathological behavior of the local optimal solution and improve the model prediction ability.

[0022] Furthermore, under a constant heating rate, the expression of the DAEM equations for the pseudo-component j at temperature T is as follows:

[0023]

[0024]

[0025] Where n, A j , E i , R and T are the reaction order, pre-exponential factor, activation energy, ideal gas constant and instantaneous absolute temperature at time t, respectively, and α j represents the mass fraction of pyrolysis volatiles of pseudo component j.

[0026] Furthermore, assuming that there is no mutual influence between the pseudo-components j, the overall pyrolysis reaction can be considered as a linear combination of the decomposition reactions of the pseudo-components, expressed as follows:

[0027]

[0028]

[0029] Where M represents the number of pseudo components that need to be calculated.

[0030] Since the DAEM equations contain two layers of integration, namely the inner temperature integral and the outer activation energy integral, as shown in equations (1)-(2), we cannot obtain the exact analytical solution of the equations, and the numerical solution is also extremely complicated.

[0031] Therefore, in order to save program computing time and memory, a sixth-order Padé approximation equation is introduced to simplify the inner layer temperature integral, as shown below. When x>12, the absolute error of the sixth-order Padé approximation equation is less than 1×10-16 .

[0032]

[0033]

[0034] For wood decomposition DAEM containing three pseudo-components, the TG and DTG prediction models under thermogravimetric analysis can be established based on the above equations (1)-(6). Therefore, the DAEM parameter group can be solved according to the test results.

[0035] Compared with the prior art, the method for measuring the age of ancient building wood based on the distributed activation energy model described in the present invention has the following advantages:

[0036] The ancient building wood age measurement method based on the distributed activation energy model described in the present invention can realize the rapid measurement of the wood age, solves the shortcomings of the ancient building wood age measurement means, and thus makes up for the shortcomings of the existing wood age measurement technology such as large number of samples, long time, complex operation and high cost. Compared with the traditional method, this method is more suitable for application and promotion, can effectively improve the measurement efficiency and accuracy, and has great application value and market potential. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings:

[0038] Figure 1 It is a flow chart of the method for measuring the age of ancient building wood based on the distributed activation energy model according to an embodiment of the present invention;

[0039] Figure 2 A flowchart of combining a genetic algorithm (GA) and a particle swarm algorithm (PSO) according to an embodiment of the present invention;

[0040] Figure 3 Experimental curves of cypress at different heating rates and the curves and component contents predicted by the distributed activation energy model.

[0041] Figure 4 Experimental curves of old cypress at different heating rates and the curves and component contents predicted by the distributed activation energy model.

[0042] Figure 5 Experimental curves of pine wood at different heating rates and the curves and component contents predicted by the distributed activation energy model.

[0043] Figure 6 Experimental curves of old pine wood at different heating rates and the curves and component contents predicted by the distributed activation energy model. DETAILED DESCRIPTION

[0044] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0045] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0046] In order to measure the age of wood in components of ancient buildings, handicrafts, etc., the present invention combines experimental testing, theoretical modeling and program operation to develop a method for measuring the age of ancient building wood based on a distributed activation energy model.

[0047] Through the aging study of various woods, it was found that the lignin component in the wood is not easy to degrade as the time of exposure of the wood to the air increases, but hemicellulose and cellulose will slowly degrade over time. Therefore, if the three components in the wood can be separated, different wood ages can be distinguished according to the content change of holocellulose. The predecessors used chemical reagents to isolate, purify and extract chemical components in scientific research, and quantitatively studied the chemical components of wood, including moisture, extracts, ash, holocellulose (cellulose and hemicellulose) by the difference of chemical weight. Although this method can obtain the content change of cellulose components, the limitations of large sampling amount, long test cycle and complex operation make its application in the present invention less valuable. Therefore, the present invention intends to analyze the thermogravimetric results by microsampling the sample to be tested and carrying out thermogravimetric analysis tests under multiple heating rates, develop DAEM multi-parameter optimization program based on the distributed activation energy model, obtain the kinetic parameters of multiple components, and judge the age of the wood to be tested.

[0048] Wood pyrolysis is a complex process that produces a large number of intermediates and consists of multiple parallel reaction processes. The distributed activation energy model deconvolutes the entire total package decomposition process by assuming a set of parallel and independent reaction processes. Each reaction process is considered to be composed of a series of related "small" reaction groups, assuming that volatiles are generated by the reaction of pseudo-species in a pseudo-component, and each pseudo-component corresponds to several unknown real chemicals. Furthermore, we can assume that the decomposition process of a pseudo-component is the sum of the decomposition of a series of independent pseudo-chemical components. For the wood pyrolysis reaction, it can be considered as a combination of three components (corresponding to hemicellulose, cellulose and lignin, respectively). For pseudo-component j, it can be considered that it has many chemical components (corresponding to pseudo-components) that undergo decomposition reactions at the molecular level, all of which conform to the nth-order reaction rate equation, and it is assumed that the component population it contains conforms to a set of continuous activation energy distribution rules.

[0049] At this point, we can introduce a probability density function f(E) to describe the distribution of activation energy of the component group. After model derivation, the DAEM equations for pseudo-component j at temperature T at a constant heating rate are established, as shown in equations (1)-(2). After simplification, the results shown in equations (5)-(6) are obtained.

[0050] For wood decomposition DAEM containing three pseudo components, the TG and DTG prediction models under thermogravimetric analysis were established based on equations (1)-(6).

[0051] Therefore, the DAEM parameter group is solved according to the test results. In order to avoid the interference of the dynamic compensation effect on the DAEM parameter group solution process, the solution result shows the "pathological behavior" of the local optimal solution. The present invention solves α and dα / dT under multiple heating rates, increases the amount of data and expands the similarity between parameters, and constructs a DAEM multi-parameter optimization program based on the MATLAB platform.

[0052] In addition, the present invention uses the regression sum of squares (SSR) as a measure of fit to evaluate the fitting ability of the DAEM model. The specific content is the sum of the squares of the difference between the thermogravimetric predicted data point and the experimental data point and the squares of the difference between the DTG predicted data point and the experimental data point at different heating rates. It is worth noting that the normalized TGA value range is between 0 and 1, and the DTG data obtained by first-order differentiation of the TGA curve is of a lower order of magnitude.

[0053] Therefore, in order to more intuitively compare the differences between experimental data and predicted data, we set the coefficients 10 (the square of the difference between TGA data points) and 100 (the square of the difference between DTG data points) in the formula. From a statistical point of view, the lower the SSR of the model, the stronger its ability to fit the actual data points. Its objective function is shown in the following formula:

[0054]

[0055] In the above formula, N β and N s are the heating rate and the number of selected experimental data, respectively. The subscripts β and s represent the experimental heating rate and the number of selected experimental points, respectively. The subscripts exp and fit represent the experimental data and the theoretical number, respectively. The kinetic parameters of the three pseudo-components can be obtained by solving the program.

[0056] Furthermore, the parameters and DTG curves corresponding to the three groups of pseudo-components obtained by DAEM, combined with the specific pyrolysis temperature ranges of lignin, cellulose and hemicellulose, can easily identify the pseudo-component parameters corresponding to the three components.

[0057] The study found that for a given wood species, there is a corresponding relationship between the kinetic parameters (activation energy, pre-exponential factor and weight) of the three components and the aging time. Therefore, we can obtain a database of the corresponding wood age-kinetic parameters by artificially aging the wood to be tested. According to the kinetic parameters of the wood to be tested, the corresponding wood age can be obtained by comparing the database.

[0058] Example

[0059] A method for measuring the age of ancient building wood based on the distributed activation energy model includes sample preparation, thermogravimetric testing, DAEM parameter calculation, database establishment and determination of wood age composition, such as Figure 1 A flowchart of the method is shown in FIG.

[0060] (1) Sample preparation:

[0061] Firstly, four samples of cypress, pine and old wood were selected. The samples were all taken from Sichuan, China. The naturally aged wood came from the discarded wood of local buildings in the late Qing Dynasty, and the new wood was taken from local freshly felled wood.

[0062] In order to select the part of the wood to be tested that is not easily susceptible to corrosion and pollution, and not to cause damage to wooden components or crafts, after removing surface dust and impurities, use a tubular shield to surround the sampling site to prevent dust and other contamination during the sampling process, and then use an electric drill to drill out about 50 mg of the sample to be tested, ball mill the sample to be tested, and pass it through a 40-60 mesh sieve to ensure uniform powder particles. After drying in an oven at 105°C for 24 hours, take out and prepare thermogravimetric test samples, and each sample is kept at 3 mg.

[0063] (2) Thermogravimetric test:

[0064] In order to avoid the dynamic compensation effect causing the solution to fall into the local optimal solution, it is preferred to solve the α and dα / dT data at three or four different heating rates simultaneously to obtain a more reliable DAEM optimal parameter group.

[0065] In this example, 10, 15 and 20 K·min were selected -1 For these three heating rates, the heating range is preferably from room temperature to 850°C.

[0066] Thermogravimetric testing is to evenly cover the sample to be tested on the bottom of the crucible, and use a thermogravimetric analyzer or a synchronous thermal analyzer to perform thermogravimetric testing at different heating rates. After each group of tests is completed, the relevant results are sorted out, and the TG and DTG curves of the sample to be tested at different heating rates are plotted. If the curves are inconsistent, analyze the relevant reasons and repeat the test.

[0067] like Figure 3-Figure 6 Figure a shows the selected cypress, old cypress, pine and old pine at 10, 15 and 20 K·min -1From the TG / DTG curves under the heating rate, it can be observed that the main weight loss process of the four woods occurs between 500-650K, and after 650K, the mass shows a trend of slowly decreasing. The DTG curves have peaks around 600K, indicating that the temperature corresponding to the maximum pyrolysis rate of the four woods is around 600K. In addition, there is an obvious shoulder peak between 500-600K, and there are "bulge" peaks between 620-800K.

[0068] like Figure 3-Figure 6 Figure b shows the four wood materials at 10K·min -1 The DTG experimental value obtained below is compared with the predicted curve calculated by the DAEM model. It can be seen from the figure that the predicted DTG data can match the experimental data very well, which shows that the DAEM program developed based on the distributed activation energy model can well predict the pyrolysis process of the four types of wood.

[0069] Figure 3-Figure 6 Figure c shows the distribution of three pseudo-components during the pyrolysis of different wood species obtained by running the DAEM program, which correspond to the pyrolysis of the three main components in the wood (hemicellulose, cellulose and lignin), which well demonstrates the superiority of the program calculation and accurately separates the pyrolysis process of hemicellulose, cellulose and lignin. Hemicellulose is a non-cellulose polysaccharide with a low degree of polymerization in wood tissue. It is the main component of plant cell walls and is the first to pyrolyze among the three components; cellulose has a high content in wood and has 5 different crystal forms with good thermal stability. The pyrolysis of lignin almost participates in the entire pyrolysis process of wood, which is also shown as a broad peak in the DTG curve. The experimental results also show that wood will not lose any of its main components after the natural aging process, which also shows the feasibility of judging the age of wood by the kinetic parameters of the three main components.

[0070] Figure 3-Figure 6 Figure d shows the differences in the contents of the three main components of cypress, pine and their natural aging. The results show that the aged wood all showed a decrease in hemicellulose content, but a relatively increased content of cellulose and lignin.

[0071] (3) DAEM parameter calculation:

[0072] The DAEM parameter calculation refers to a DAEM parameter group solving program developed based on a multi-parameter optimization algorithm, preferably a coupling of a particle swarm algorithm (PSO) and a genetic algorithm (GA), which can better obtain a global optimal solution. The DAEM model preferably has three pseudo components, namely lignin, cellulose and hemicellulose, and the activation energy distribution model f(E) preferably has Gaussian distribution, logistic distribution and Weibull distribution.

[0073] In this specific implementation, Weibull distribution is selected as the activation energy distribution model f(E).

[0074] Its probability density function is as follows:

[0075]

[0076] E0=ηΓ(1+1 / λ)+γ (9)

[0077]

[0078]

[0079] Where λ>0 is the shape parameter, η>0 is the scale parameter of the distribution, and γ is the activation energy threshold, which satisfies equation (8) only when E>γ. The average activation energy E0 and the standard deviation σ are defined as equations (9) and (10), where Γ is the gamma function, which is defined by a convergent improper integral for real numbers x>0.

[0080] It should be introduced that in the operation process of the DAEM model, we set each pseudo-component to contain 4 variables as optimization parameters, namely ln A (pre-exponential factor), σ (standard deviation as weight factor), E0 (average activation energy) and c (component content). The set multiple parameters are optimized through the algorithm written. The algorithm is based on the optimization method of combining GA (genetic algorithm) and PSO (particle swarm optimization) algorithms to solve the DAEM problem in order to achieve the purpose of mutual complementation and verification. Figure 2 The specific operation process of DAEM is shown in the figure.

[0081] When performing the operation, we need to set the maximum number of evolutionary generations and randomly generate 12 individuals as the initial population; evaluate and select the fitness of each individual based on SSR, and perform crossover and iteration. Next, after 1000 iterations, select a population with the highest fitness as the initial value of the next process. The second step is to bring the obtained optimal population into the PSO algorithm for testing and optimization again, with the aim of obtaining the global optimal parameters. Before using the PSO method, we specified the variable range of each parameter (lnA: 10~60; σ: 0~70; E a :80~350kJ mol -1 ; c: 0~60%. ), the best parameters can be found within this range, and the best parameters are output and listed in Table 1 at the end of the program.

[0082] Table 1. Pre-exponential factors lnA, σ weight factors, activation energy E0 and component contents c of three groups of pseudo-components of different wood species calculated based on the distributed activation energy model

[0083]

[0084] Table 1 lists the optimal parameter groups for cypress, old cypress, pine and old pine solved by the DAEM parameter group. It can be seen that the pre-exponential factors and activation energies of each pseudo-component are different for different tree species and wood species with different aging times, and the contents of the three main components of the same wood species are different before and after aging.

[0085] (4) Establish database:

[0086] The above operation is performed on the wood to be tested with known years of artificial aging, and a database of the corresponding relationship between time and kinetic parameters is calculated and established;

[0087] (5) Determine the age of the wood:

[0088] For unknown wood to be tested, select the part of the wood to be tested that is not corroded and will not cause damage to the sample, remove dust and impurities on the surface of the sample, and use a tool to cut or drill the sample to be tested; repeat the above steps to calculate the kinetic parameters of each pseudo-component, namely the activation energy, pre-exponential factor and weight factor of lignin, cellulose and hemicellulose, and determine the age of the wood to be tested by comparing with the established database.

[0089] The present invention solves the shortcomings of the measurement methods for the wood age of ancient buildings. Compared with traditional methods, this method is more suitable for application and promotion, can effectively improve measurement efficiency and accuracy, and has great application value and market potential.

[0090] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for measuring the age of ancient building wood based on a distributed activation energy model, characterized in that: The method comprises the following steps: (1) Sample preparation: Select the wood samples of known years obtained from the ancient building complex, grind them into powder, sieve them, dry them, and weigh them in equal parts for later use; (2) Thermogravimetric test: Carry out thermogravimetric tests on the samples at different heating rates, draw TG and DTG curves at different heating rates, and analyze the reliability of the test data; (3) DAEM parameter calculation: Based on the thermogravimetric results at multiple heating rates, a distributed activation energy prediction model for n-order reactions containing three pseudo-components was established. The developed DAEM multi-parameter optimization program was used to solve the problem, obtain the optimal DAEM parameter group for the sample to be tested, and determine the kinetic parameters of each pseudo-component. The developed DAEM multi-parameter optimization program is based on the coupling of particle swarm optimization and genetic algorithm, and solves α and dα / dT data under multiple groups of heating rates at the same time, avoiding the pathological behavior of local optimal solutions and improving the model prediction ability; Under constant heating rate, the expression of DAEM equations for pseudo component j at temperature T is as follows: Where n, A j , E i , R and T are the reaction order, pre-exponential factor, activation energy, ideal gas constant and instantaneous absolute temperature at time t, respectively, and α j represents the mass fraction of pyrolysis volatiles of pseudo component j; (4) Establish database: Repeat the above steps (1)-(3) to calculate the kinetic parameters of each component of wood of the same species but different years, so as to establish a database of the corresponding relationship between time and kinetic parameters of wood of known years; (5) Determine the age of the wood: Select a part of the wood to be tested that is not contaminated by corrosion and does not cause damage to the sample, remove dust and impurities on the surface of the sample, and use a tool to cut or drill the sample to be tested; then repeat the above steps (1)-(3) to calculate the kinetic parameters of each component, and determine the age of the wood to be tested by comparing it with the established database; Among them, the three pseudo-components refer to lignin, cellulose and hemicellulose respectively; the kinetic parameters refer to the activation energy, pre-exponential factor and weight factor of lignin, cellulose and hemicellulose.

2. The method for measuring the age of ancient building wood based on the distributed activation energy model according to claim 1 is characterized in that: After ball milling in step (1), the product is sieved through a 40-60 mesh sieve and dried in an oven at 75-105° C. for 24-48 hours. Each sample is weighed to 3-5 mg for later use.

3. The method for measuring the age of ancient building wood based on the distributed activation energy model according to claim 1 is characterized in that: In step (2), select 3-4 heating rates and set the heating range to 25-850°C.

Images