Method, device, equipment, medium and product for rapidly determining chemical properties of larch wood

Abstract

The invention discloses a method, a device, equipment, a medium and a product for rapidly measuring chemical properties of larch wood, and relates to the field of forestry science and spectral analysis, the method comprises the following steps: collecting a near infrared spectrum of a larch sample in a preset wavelength range, and sequentially selecting an effective modeling spectrum region and pre-treating; determining the lignin content, the alpha-cellulose content and the holocellulose content of the larch sample as reference values; constructing a training set and a prediction set; establishing a near infrared spectrum prediction model of lignin, alpha-cellulose and holocellulose; verifying the prediction precision of the near infrared spectrum prediction model of lignin, alpha-cellulose and holocellulose by using the prediction set; inputting a larch sample to be detected into the model which is verified to be effective, and outputting the predicted content of lignin, alpha-cellulose or holocellulose. The method realizes nondestructive, rapid and high-precision detection of main chemical components of larch wood, and is suitable for forest tree genetic breeding and wood quality evaluation.

Description

Technical Field

[0001] This application relates to the fields of forestry science and spectral analysis, and in particular to a method, apparatus, equipment, medium, and product for rapid determination of the chemical properties of larch wood. Background Technology

[0002] Japanese larch (Larix kaempferi) is a deciduous tree native to the mountainous regions of Honshu Island in central Japan. This species is highly adaptable and widely cultivated, possessing significant ecological and economic value, particularly in timber production and pulp and paper manufacturing (Park and Fowler, 1982; Takata et al., 2005). Over a century of introduction and cultivation in China has shown that this species exhibits remarkable rapid growth, generally outperforming native larch species. It has now become an important coniferous afforestation species in the high-altitude regions of Northeast, North, Northwest, and Southwest China (Ma, 2008).

[0003] Lignin and cellulose are the main components of wood and have a significant impact on wood property formation. Cellulose accounts for approximately 40-50% of the total volume of wood and is mainly composed of β-D-1,4-glucosidic bonds. It is the main skeletal structure of the cell wall, providing mechanical support for cells and the entire plant (Delmer et al., 1995). Lignin is mainly deposited in certain specialized cells such as vessels, sclerenchyma, and phloem fibers of the cell wall, covalently bound to hemicellulose to increase the strength and stiffness of the cell wall (Brodeur-Campbell et al., 2006; Desmond et al., 2011). The chemical components of wood, such as lignin and cellulose, have a significant impact on wood property formation and industrial applications (Bose et al., 2009). Higher lignin content significantly affects pulp yield and paper quality in papermaking (Michell et al., 1998).

[0004] The chemical properties of wood are an important aspect of wood quality improvement. Traditional methods for detecting these properties are time-consuming, labor-intensive, and costly (Fengel et al., 1989). However, near-infrared spectroscopy (NIR) non-destructive testing technology, developed in the 1950s (McClure. 2003), offers advantages such as speed, sensitivity, and high accuracy, enabling rapid testing of batches of samples and finding wide application in wood physicochemical property testing (Satoru et al., 2015). A combined NIR and partial least squares (PLS) method was used to establish wood physicochemical property testing models for subtropical pine, eucalyptus, loblolly pine, radiata pine, and willow, demonstrating high predictive accuracy.

[0005] While the aforementioned traditional testing methods are accurate, they are cumbersome, time-consuming, require large amounts of chemical reagents, and are destructive, making it difficult to meet the needs of large-scale forest tree breeding and rapid screening of timber quality.

[0006] Near-infrared spectroscopy (NIR-PLS) technology has been widely used for the detection of physicochemical properties of wood due to its advantages such as speed, non-destructive nature, and batch processing capability. Previous studies have established NIR-PLS models in eucalyptus, sago palm, and willow species to predict cellulose or lignin content. The model RL... 2 The values ​​are typically between 0.85 and 0.90. However, significant differences exist in cell wall structure, chemical composition ratios, and spectral responses among different tree species, making direct transfer from general models difficult. Currently, there is no systematic research or application of a dedicated NIR-PLS chemical trait prediction model for Japanese larch, an important afforestation tree species.

[0007] Furthermore, existing technologies often focus on model accuracy verification while neglecting the collaborative optimization of the entire process from spectral acquisition, spectral region selection, preprocessing strategies to model construction, resulting in insufficient model generalization ability. Therefore, there is an urgent need to develop a rapid measurement method specifically for Japanese larch that is complete in process, reliable in accuracy, and engineerable. Summary of the Invention

[0008] The purpose of this application is to provide a method, apparatus, equipment, medium and product for rapid determination of the chemical properties of larch wood. By systematically optimizing the near-infrared spectral acquisition, spectral region screening, data preprocessing and PLS modeling process, it is possible to achieve high-precision non-destructive prediction of lignin, α-cellulose and holocellulose content.

[0009] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a rapid method for determining the chemical properties of larch wood, including: Near-infrared spectra of larch samples were collected within a preset wavelength range; The near-infrared spectrum within the preset wavelength range is limited to a preset band as the effective modeling spectral region; The spectral data of the effective modeling spectral region are preprocessed; The lignin content, α-cellulose content, and holocellulose content of the larch samples were determined as reference values. The preprocessed spectral data is paired with the reference values ​​to construct a training set and a prediction set. Based on the training set, near-infrared spectral prediction models for lignin, α-cellulose, and holocellulose were established respectively. The prediction accuracy of the near-infrared spectral prediction models for lignin, α-cellulose, and holocellulose is verified using the prediction set. When the verification is passed, the model is deemed effective. Input the larch sample to be tested into a validated model, and output the predicted content of lignin, α-cellulose, or holocellulose.

[0010] Optionally, based on the training set, near-infrared spectral prediction models for lignin, α-cellulose, and holocellulose are established respectively, specifically including the following steps: Get One sample Concentration matrix of each component ; Obtain instrument measurements One sample Absorbance matrix at each wavelength point ; The concentration matrix and the absorbance matrix are decomposed into eigenvector form: ; ;in, and They are respectively OK The concentration characteristic factor matrix and absorbance characteristic factor matrix of the column, for order absorbance loading matrix, for Concentration loading matrix of order, and They are respectively , Concentration residual matrix and absorbance residual matrix; Based on the correlation decomposition of eigenvectors and Establish a regression model: ;in, , for A diagonal regression coefficient matrix of dimension 1. It is a random error matrix.

[0011] Optionally, the preset wavelength range is 350nm to 2500nm; the preset band range is 1006nm to 2491nm.

[0012] Optionally, based on the training set, near-infrared spectral prediction models for lignin, α-cellulose, and holocellulose are established as follows: Based on the training set, near-infrared spectral prediction models for lignin, α-cellulose, and holocellulose were established using partial least squares method.

[0013] Optionally, the prediction accuracy of the near-infrared spectral prediction models for lignin, α-cellulose, and holocellulose can be verified using the prediction set, specifically through the coefficient of determination. Correction standard deviation Cross-training set standard error Prediction standard deviation Sum of standard deviations and performance rate The value is then verified for accuracy.

[0014] Optionally, the determination coefficient Correction standard deviation Cross-training set standard error Prediction standard deviation Sum of standard deviations and performance rate The expressions for the values ​​are as follows: ; ; ; ; ; ; in, Indicates the measured value. This represents the near-infrared predicted value. This represents the average of the measured values. Represents the total number of samples. Indicates the first One sample.

[0015] Secondly, this application provides a rapid device for determining the chemical properties of larch wood, comprising: The near-infrared spectroscopy acquisition module is used to acquire the near-infrared spectrum of larch samples within a preset wavelength range; The effective modeling spectral region determination module is used to limit the near-infrared spectrum within the preset wavelength range to a preset band as the effective modeling spectral region; The preprocessing module is used to preprocess the spectral data of the effective modeling spectral region; The chemical composition determination module is used to determine the lignin content, α-cellulose content, and holocellulose content of the larch sample as reference values. The training set and prediction set construction module is used to pair the preprocessed spectral data with the reference value to construct the training set and prediction set; The prediction model building module is used to establish near-infrared spectral prediction models for lignin, α-cellulose, and holocellulose based on the training set. The verification module is used to verify the prediction accuracy of the near-infrared spectral prediction model of lignin, the near-infrared spectral prediction model of α-cellulose, and the near-infrared spectral prediction model of holocellulose using the prediction set. When the verification is passed, the model is determined to be effective. The prediction module is used to input the larch sample to be tested into a validated model and output the predicted content of lignin, α-cellulose, or holocellulose.

[0016] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method for rapid determination of the chemical properties of larch wood as described above.

[0017] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the rapid determination method for chemical properties of larch wood described above.

[0018] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the rapid determination method for chemical properties of larch wood described above.

[0019] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a method, apparatus, equipment, medium, and product for rapid determination of the chemical properties of larch wood, which has the following significant advantages: High precision: The established model R 2 All values ​​were ≥0.898, and all RPD values ​​were >2.8, meeting the accuracy requirements for wood property testing in forest tree breeding. Rapid and non-destructive: Single sample testing time is less than 5 minutes, no chemical reagents are required, and sample integrity is protected; Process standardization: A complete technology chain is formed from sampling to prediction, which facilitates promotion and application; Component-specific optimization: Differentiated preprocessing is used for different chemical components, which significantly improves the model's specificity and stability; High applicability: Covers different geographical sources and sampling sites, enhancing model robustness; Clearly specific: It is clearly limited to larch trees to avoid misuse of the model. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A flowchart illustrating a rapid method for determining the chemical properties of larch wood, provided in an embodiment of this application; Figure 2 This is a schematic diagram illustrating the correlation analysis between predicted and measured values ​​in an embodiment of this application; Figure 3 This is a schematic diagram of the functional modules of a rapid chemical property determination device for larch wood provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0022] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0024] In one exemplary embodiment, such as Figure 1 As shown, a rapid method for determining the chemical properties of larch wood is provided. This method is executed by a computer device, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method includes the following steps: Step 101: Collect the near-infrared spectrum of the larch sample within the preset wavelength range.

[0025] First, sample the test material: The NIR-PLS modeling materials consisted of two parts: disc samples and core samples (Table 1). Disc samples were collected from 28 Japanese larch trees in Hubei, Henan, and Liaoning provinces. Three 2-cm thick discs were taken from each tree at heights of 0.5 m, 1.3 m, and 1.5 m from the top, for a total of 84 discs. Core samples were collected from 18 trees in Liaoning province. Core samples were drilled from the diameter at breast height (DBH) of each tree using a 45 cm long, 12 mm diameter growth cone, running north-south. All samples were air-dried at room temperature, ground into powder, and sieved through a 40-mesh sieve. NIR spectral data were collected for each sample, and the lignin, α-cellulose, and holocellulose contents of each sample were determined using a wet chemical method.

[0026] Table 1 Basic Information of NIR-PLS Modeling Samples

[0027] Then, the contents of lignin, α-cellulose, and holocellulose were determined: Method for determining holocellulose: Place the sample in a 250ml Erlenmeyer flask, add 65ml distilled water, 0.5ml glacial acetic acid, and 0.6g sodium chlorite, shake well, cover the flask, and heat in a 75℃ constant temperature water bath for 1 hour. During heating, the Erlenmeyer flask should be rotated and shaken frequently. After 1 hour, the solution does not need to be cooled. Add another 0.5ml glacial acetic acid and 0.6g sodium chlorite, shake well, and continue heating in the 75℃ water bath for another hour. Repeat this process until the sample turns white. Remove the Erlenmeyer flask from the water bath and cool it in an ice-water bath. Filter the solution using a pre-weighed glass filter, wash repeatedly with distilled water until the filtrate is no longer acidic, and finally wash three times with acetone. Blot dry the filtrate, remove the filter, and wash the outside of the filter with distilled water. Place it in a drying oven and dry at 105℃ until constant weight. Compare the dry weight after drying with the oven-dry weight of the sample to obtain the wood palm cellulose content.

[0028] α-cellulose determination method: Weigh 1g of palm cellulose sample and place it in a 200ml beaker. Add 25ml of sodium hydroxide (17.5%) and shake. After 16 minutes, add 25ml of distilled water and shake for 5 minutes. Then filter through a glass filter crucible of known weight, rinse with distilled water until neutral, add 40ml of 10% acetic acid, heat for 5 minutes, and weigh the residue. This residue is the α-cellulose content.

[0029] Lignin determination method: After crushing the wood core and passing it through a 200-mesh sieve, weigh 0.1g and place it in a centrifuge tube. Add 10ml of 1% acetic acid, shake well, and centrifuge. Wash the precipitate once with 5ml of 1% acetic acid, then add 4ml of a mixture of ethanol and ether, soak for 3 minutes, and wash 3 times. Evaporate the precipitate to dryness in a boiling water bath. Then add 3ml of 72% sulfuric acid to the precipitate, shake well, and let it stand at room temperature for 16 hours to allow the cellulose to dissolve completely. Then add 10ml of distilled water to the test tube, shake well, place in a boiling water bath for 5 minutes, cool, add 5ml of distilled water and 0.5ml of 10% barium chloride solution, shake well, and centrifuge. After precipitation, rinse twice with distilled water. Then add 10 ml of 10% sulfuric acid and 10 ml of 0.025 mol / L potassium dichromate solution to the washed lignin precipitate. Place the test tube in a boiling water bath for 15 minutes, stir and cool. Pour all the substances in the test tube into a beaker for titration. Add 5 ml of 20% KI solution and 1 ml of 1% starch solution. Measure the lignin content by titration with sodium thiosulfate. At the same time, perform a reagent blank test.

[0030] Finally, NIR spectral scanning was performed using a LabSpec® Pro near-infrared spectrometer (ASD, USA), with a detector of 350 nm–2500 nm and a low-noise 512-element PDA. Spectral sampling intervals were 1.4 nm in the 350 nm–1050 nm band and 2 nm in the 1000 nm–2500 nm band. The spectral resolution was 3 nm at a working wavelength of 700 nm and 10 nm at working wavelengths of 1400 nm and 2100 nm. A cup light source monitor was used for blank calibration of a commercial PTFE white board. The full spectrum of Japanese larch wood powder in the 350 nm–2500 nm range was acquired using Indico software (ASD). Each sample was scanned three times, and the average of the 30 spectral points from each scan was taken as one spectral value. The average of the three spectral values ​​for each sample was then calculated using Unscrambler 9.2 software (CAMO, Sweden). The probability of each sample's spectral data deviating from the spectral cluster center was calculated using the Mahalanobis distance algorithm in Foss NIRSystems version 2.51 software. Samples with a probability level exceeding 0.95 were discarded. The Euclidean distance between samples was then calculated using the Euclidean distance algorithm (Yeh et al., 2005). Step 102: Limit the near-infrared spectrum within the preset wavelength range to a preset band as the effective modeling spectral region.

[0031] When using spectral information for modeling, selecting a spectral region that is too wide will introduce a lot of invalid information, while selecting a region that is too narrow will reduce the amount of valid information. Therefore, optimizing the spectral region is a crucial step in establishing a high-accuracy prediction model. This application preliminarily analyzed the near-infrared spectral information of the samples and found that the near-infrared spectral information in the short-wavelength region of 350 nm to 1000 nm was weak and did not improve the modeling, while the spectral data in the 1000 nm to 2500 nm range was richer. To obtain the optimal spectral region, this application first calculated the regression coefficient distribution of absorbance at each wavelength point of the full spectrum with the content value of the analyzed reference value. A smooth spectral region in the 1006 nm to 2486 nm band showed a high correlation with the chemical properties of wood; therefore, this band was selected as the modeling spectral region for further spectral data preprocessing.

[0032] Step 103: Preprocess the spectral data of the effective modeling spectral region.

[0033] To eliminate the influence of optical path inconsistencies and noise signals during near-infrared spectroscopy measurements, this application preprocessed the spectral data for a selected spectral region. Appropriate processing methods were selected by analyzing the values ​​of R², SEC, and SECV using derivatives and smoothing techniques. 2 The closer the values ​​are to 1, the smaller the SEC and SECV values, and the better the model's predictive performance. In this case, the data processing method is optimal. After comparing the values ​​of R², SEC, and SECV, the optimal prediction results were found for α-cellulose using the original spectrum from 1006 nm to 2491 nm; for holocellulose, the optimal prediction results were found for data processed using the first derivative of the 1006 nm to 2486 nm spectrum with 15 points of smoothing; and for lignin content determination, the optimal prediction results were found for data processed using the first derivative of the 1006 nm to 2491 nm spectrum with 15 points of smoothing.

[0034] Step 104: Determine the lignin content, α-cellulose content, and holocellulose content of the larch sample as reference values.

[0035] Step 105: Pair the preprocessed spectral data with the reference values ​​to construct a training set and a prediction set.

[0036] Step 106: Based on the training set, establish near-infrared spectral prediction models for lignin, α-cellulose, and holocellulose, respectively.

[0037] Establishing a calibration model using a multivariate calibration method is a crucial step in near-infrared spectroscopy quantitative analysis. Commonly used modeling methods include multiple linear regression, principal component analysis, principal component regression, partial least squares (PLS), support vector machines, and artificial neural networks. Among these, partial least squares (PLS) combines the advantages of ordinary least squares (OLS) and principal component analysis, simultaneously extracting characteristic wavelength variables and establishing a regression model. By performing different linear combinations on the original wavelength independent variable matrix, new variables are constructed that are linearly represented by the original variables and are mutually independent, while retaining as much useful information as possible. This also ensures that the correlation between the new variables and the dependent variable matrix is ​​maximized. Therefore, this method not only eliminates collinearity among the original independent variables but also ensures that the established regression model fully reflects the relationship between the independent and dependent variables. This application uses partial least squares (PLS) to construct a near-infrared prediction model for the chemical properties (α-cellulose, lignin, and holocellulose) of Japanese larch wood. The specific steps for constructing the model using partial least squares (PLS) are as follows: First, we... (102) samples (3) Concentration matrix of the components and instrument measurement (102) samples Absorbance matrix at (1501) wavelength points Decompose into eigenvector form: ; (in and They are respectively OK List( The concentration characteristic factor matrix and absorbance characteristic factor matrix (for abstract components) for order absorbance loading matrix, for Concentration loading matrix of order, and They are respectively , (Concentration residual matrix and absorbance residual matrix). Based on the correlation decomposition of eigenvectors. and Establish a regression model: ( for A diagonal regression coefficient matrix of dimension 1. (For random error matrix). If the absorbance vector of the sample to be tested is x The concentration is .

[0038] Step 107: Validate the prediction accuracy of the near-infrared spectral prediction models for lignin, α-cellulose, and holocellulose using the prediction set. If the validation is successful, the model is deemed effective.

[0039] Using 70% of the samples as the training set and 30% as the prediction set (Table 2), NIR-PLS models for α-cellulose, lignin, and holocellulose were established using partial least squares (PLS1). Finally, the coefficient of determination (R²) was used to... 2 The model's predictive accuracy is evaluated using the corrected standard deviation (SEC), cross-training set standard error (SECV), predicted standard deviation (SEP), and standard deviation performance ratio (RPD) (Yoshio et al., 2010). The coefficient of determination R0 2 The closer the RPD value is to 1, the better the model fits; the smaller the correction standard deviation, prediction standard deviation, and cross-training set standard error, the better the model fits. When analyzing forestry experimental materials, an RPD value of around 1.5 can be used for a rough detection of wood chemical properties; when the RPD value is greater than 2.5, the model accuracy is high and can fully meet the detection accuracy requirements for wood properties in breeding.

[0040] ; ; ; ; ; ; in, Indicates the measured value. This represents the near-infrared predicted value. This represents the average of the measured values. Represents the total number of samples. Indicates the first One sample.

[0041] Step 108: Input the larch sample to be tested into the validated model, and output the predicted content of lignin, α-cellulose or holocellulose.

[0042] This application analyzes the near-infrared spectral characteristics of lignin, α-cellulose, and holocellulose in Japanese larch wood. Through screening of the effective spectral region, and after smoothing the derivatives of the effective spectral region, it uses the R-squared value of the model. 2Following the selection principle that higher RPD values ​​and lower SEC and SEP values ​​result in higher model prediction accuracy, near-infrared spectral prediction models for lignin, α-cellulose, and holocellulose were ultimately established using partial least squares method. Descriptive statistical results for the three models in this application are shown in Table 2. In the training and prediction sets of the models, excluding α-cellulose, the RPD values ​​are... 2 Value (R) 2 Apart from being slightly lower at 0.898, the R values ​​of the other two components were... 2 All values ​​were greater than 0.9; SECV values ​​were 0.554, 0.840, and 0.611, and SEC values ​​were 0.560, 0.850, and 0.620, respectively. α-cellulose had a slightly higher SEP of 1.10, while the other two components had lower values ​​of 0.620 and 0.840, respectively. SEC and SECV values ​​were similar, but SEP values ​​were slightly higher than SEC and SECV values. Further correlation analysis between predicted and measured values ​​was performed (see details). Figure 2 , Figure 2 The correlation between predicted and measured NIR values ​​in the training sets (parts (a), (c), and (e)) and prediction sets (parts (b), (d), and (f)) for three wood chemical components is shown. Parts (a) and (b) represent α-cellulose; parts (c) and (d) represent lignin; and parts (e) and (f) represent holocellulose. The results show that the RPD values ​​of the three models are relatively high, at 2.920, 2.800, and 3.960, respectively, indicating that the models have high prediction accuracy. In summary, the optimal prediction results for α-cellulose using the original spectrum from 1006 nm to 2491 nm were obtained using partial least squares (PLS). For holocellulose, the optimal prediction results were obtained by smoothing the data using the first derivative of the 1006 nm to 2486 nm spectrum over 15 points and then using PLS. Similarly, the optimal prediction results for lignin content determination were obtained by smoothing the data using the first derivative of the 1006 nm to 2491 nm spectrum over 15 points and then using PLS. Therefore, the three NIR-PLS models for wood chemical properties established in this application all possess high detection accuracy and can be widely applied to breeding research on Japanese larch.

[0043] Table 2. Descriptive statistics of the modeled samples and statistical results of the cross-validation set. Based on the same inventive concept, this application also provides a rapid chemical property testing device for larch wood to implement the rapid chemical property testing method for larch wood described above. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the rapid chemical property testing device for larch wood provided below can be found in the limitations of the rapid chemical property testing method for larch wood described above, and will not be repeated here.

[0044] In one exemplary embodiment, such as Figure 3 As shown, a rapid chemical property determination device for larch wood is provided, comprising: The near-infrared spectroscopy acquisition module 201 is used to acquire the near-infrared spectrum of larch samples within a preset wavelength range; The effective modeling spectral region determination module 202 is used to limit the near-infrared spectrum within the preset wavelength range to a preset band as the effective modeling spectral region; Preprocessing module 203 is used to preprocess the spectral data of the effective modeling spectral region; The chemical composition determination module 204 is used to determine the lignin content, α-cellulose content and holocellulose content of the larch sample as reference values. Training set and prediction set construction module 205 is used to pair the preprocessed spectral data with the reference value to construct training set and prediction set; The prediction model building module 206 is used to establish near-infrared spectral prediction models for lignin, α-cellulose, and holocellulose based on the training set. The verification module 207 is used to verify the prediction accuracy of the near-infrared spectral prediction model of lignin, the near-infrared spectral prediction model of α-cellulose and the near-infrared spectral prediction model of holocellulose using the prediction set. When the verification is passed, the model is determined to be effective. The prediction module 208 is used to input the larch sample to be tested into a validated model and output the predicted content of its lignin, α-cellulose or holocellulose.

[0045] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 4As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores data for the rapid determination of the chemical properties of larch wood. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for the rapid determination of the chemical properties of larch wood.

[0046] Those skilled in the art will understand that Figure 4 The structures shown are merely block diagrams of some structures related to the present application and do not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements. In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0047] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0048] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0049] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0050] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0051] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0052] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0053] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A rapid method for determining the chemical properties of larch wood, characterized in that, The rapid method for determining the chemical properties of larch wood includes: Near-infrared spectra of larch samples were collected within a preset wavelength range; The near-infrared spectrum within the preset wavelength range is limited to a preset band as the effective modeling spectral region; The spectral data of the effective modeling spectral region are preprocessed; The lignin content, α-cellulose content, and holocellulose content of the larch samples were determined as reference values. The preprocessed spectral data is paired with the reference values ​​to construct a training set and a prediction set. Based on the training set, near-infrared spectral prediction models for lignin, α-cellulose, and holocellulose were established respectively. The prediction accuracy of the near-infrared spectral prediction models for lignin, α-cellulose, and holocellulose is verified using the prediction set. When the verification is passed, the model is deemed effective. Input the larch sample to be tested into a validated model, and output the predicted content of lignin, α-cellulose, or holocellulose.

2. The rapid method for determining the chemical properties of larch wood according to claim 1, characterized in that, Based on the training set, near-infrared spectral prediction models for lignin, α-cellulose, and holocellulose are established, respectively, including the following steps: Get One sample Concentration matrix of each component ; Obtain instrument measurements One sample Absorbance matrix at each wavelength point ; The concentration matrix and the absorbance matrix are decomposed into eigenvector form: ; ;in, and They are respectively OK The concentration characteristic factor matrix and absorbance characteristic factor matrix of the column, for order absorbance loading matrix, for Concentration loading matrix of order, and They are respectively , Concentration residual matrix and absorbance residual matrix; Based on the correlation decomposition of eigenvectors and Establish a regression model: ;in, , for A diagonal regression coefficient matrix of dimension 1. It is a random error matrix.

3. The rapid method for determining the chemical properties of larch wood according to claim 1, characterized in that, The preset wavelength range is 350nm to 2500nm; the preset band range is 1006nm to 2491nm.

4. The rapid method for determining the chemical properties of larch wood according to claim 1, characterized in that, Based on the training set, near-infrared spectral prediction models for lignin, α-cellulose, and holocellulose are established as follows: Based on the training set, near-infrared spectral prediction models for lignin, α-cellulose, and holocellulose were established using partial least squares method.

5. The rapid method for determining the chemical properties of larch wood according to claim 1, characterized in that, The prediction accuracy of the near-infrared spectral prediction models for lignin, α-cellulose, and holocellulose was verified using the prediction set, specifically through the coefficient of determination. Correction standard deviation Cross-training set standard error Prediction standard deviation Sum of standard deviations and performance rate The value is then verified for accuracy.

6. The rapid method for determining the chemical properties of larch wood according to claim 5, characterized in that, The coefficient of determination Correction standard deviation Cross-training set standard error Prediction standard deviation Sum of standard deviations and performance rate The expressions for the values ​​are as follows: ； ； ； ； ； ； in, Indicates the measured value. This represents the near-infrared predicted value. This represents the average of the measured values. Represents the total number of samples. Indicates the first One sample.

7. A rapid device for determining the chemical properties of larch wood, characterized in that, The rapid chemical property determination device for larch wood includes: The near-infrared spectroscopy acquisition module is used to acquire the near-infrared spectrum of larch samples within a preset wavelength range; The effective modeling spectral region determination module is used to limit the near-infrared spectrum within the preset wavelength range to a preset band as the effective modeling spectral region; The preprocessing module is used to preprocess the spectral data of the effective modeling spectral region; The chemical composition determination module is used to determine the lignin content, α-cellulose content, and holocellulose content of the larch sample as reference values. The training set and prediction set construction module is used to pair the preprocessed spectral data with the reference value to construct the training set and prediction set; The prediction model building module is used to establish near-infrared spectral prediction models for lignin, α-cellulose, and holocellulose based on the training set. The verification module is used to verify the prediction accuracy of the near-infrared spectral prediction model of lignin, the near-infrared spectral prediction model of α-cellulose, and the near-infrared spectral prediction model of holocellulose using the prediction set. When the verification is passed, the model is determined to be effective. The prediction module is used to input the larch sample to be tested into a validated model and output the predicted content of lignin, α-cellulose, or holocellulose.

8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the method for rapid determination of the chemical properties of larch wood according to any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the method for rapid determination of the chemical properties of larch wood as described in any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the method for rapid determination of the chemical properties of larch wood as described in any one of claims 1-6.

Images