Method for predicting hydrothermal humification potential of lignocellulose wastes

Through the particle swarm algorithm, the least squares support vector machine model is optimized, combined with hydrothermal carbonization and alkali hydrothermal technology, a method to predict the hydrothermal humification potential of lignocellulose waste was established, and the humification efficiency in the existing technology was solved The problem of low and lack of predictive models achieves efficient humic acid yield and environmentally friendly environmental treatment.

CN119069014AActive Publication Date: 2024-12-03TSINGHUA UNIVERSITY
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202411073269.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-06
Publication Date
2024-12-03
Estimated Expiration
2044-08-06

AI Technical Summary

Technical Problem

In the prior art, lignocellulose waste has low hydrothermal humification efficiency and lacks effective prediction models, resulting in insufficient humic acid yield and environmental treatment is not environmentally friendly.

Method used

The least squares support vector machine model (PSO-LS-SVM) is optimized by using particle swarm algorithm to combine hydrothermal carbonization and alkali hydrothermal processes to establish a process parameter database to predict the humification potential of lignocellulosic waste through hydrothermal conditions.

Benefits of technology

It improves the hydrothermal humification efficiency of lignocellulose waste, provides quantitative evaluation indicators and prediction models, and is suitable for various types of lignocellulose waste and complex working conditions, saving experimental costs and energy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure FT_1
    Figure FT_1
  • Figure FT_2
    Figure FT_2
  • Figure FT_3
    Figure FT_3
Patent Text Reader

Abstract

The invention discloses a method for predicting the hydrothermal humification potential of lignocellulose wastes. The method comprises the following steps: (1) identifying a main path of hydrothermal humification of the lignocellulose waste; (2) setting quantitative evaluation indexes to divide working conditions of a main path of hydrothermal humification; (3) determining a quantitative relationship between the performance of the key reaction intermediate and the yield of the product humic acid under each working condition; (4) constructing a process parameter database for hydrothermally preparing humic acid from lignocellulose wastes through key reaction intermediates; and (5) establishing a particle swarm optimization optimization least square support vector machine (PSO-LS-SVM) algorithm by taking the hydrothermal condition as input and the humic acid yield as output, so as to predict the hydrothermal humification potential of the lignocellulose wastes.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of waste resource utilization and model algorithms, and in particular to a method for predicting the hydrothermal humification potential of lignocellulose waste. Background Art

[0002] Lignocellulosic wastes are slow to degrade in nature due to their high cellulose and lignin content. They need to undergo a series of physical, chemical and biological processes to be stored in the soil in the form of humus. It is reported that humus accounts for up to about 70% of the total organic matter content in the soil and can promote plant growth and metabolism. In addition, since humus is rich in oxygen-containing functional groups, it can be used as anion affinity agent, adsorbent, modified catalyst, etc., and has a wide range of uses. With the acceleration of the intensive development of the planting industry and the improvement of the level of urban greening, a large amount of lignocellulosic waste (such as agricultural and forestry waste, gardening waste, fruit and vegetable waste, etc.) is generated. The current mainstream treatment technologies such as landfill and incineration not only cut off the biogeochemical cycle of resources, but also pollute the environment to a certain extent. Therefore, exploring a clean and efficient method to convert lignocellulosic waste into humus has important social, environmental and economic significance.

[0003] Hydrothermal technology can directionally convert lignocellulosic waste into humus-rich products within a few hours. However, the yield of converting lignocellulosic waste into humic acid by the one-step hydrothermal method is low, less than 5%. In response to this, the inventor team proposed a two-step hydrothermal humification technology for lignocellulosic waste, which successfully broke through the bottleneck of the current low efficiency of hydrothermal humification and further optimized the process, and authorized two invention patents (ZL 2021 1 0756585.X and ZL 2022 10885763.3).

[0004] However, the hydrothermal humification pathway of lignocellulosic waste is very complex, and the humic acid yield is affected by many conditions and factors. The quantitative relationship of hydrothermal humification of lignocellulosic waste has not yet been identified, and there is a lack of relevant prediction models for the hydrothermal humification yield of lignocellulosic waste. In the current era of artificial intelligence, the present invention utilizes the data related to hydrothermal humification of lignocellulosic waste that the applicant has obtained, combined with a large number of literature on the process parameters of hydrothermal carbonization of lignocellulosic waste, to establish a particle swarm algorithm optimized least squares support vector machine model (PSO-LS-SVM) algorithm for predicting the hydrothermal humification potential of lignocellulosic waste. Summary of the invention

[0005] The purpose of the present invention is to propose a method for predicting the hydrothermal humification potential of lignocellulosic waste. The method first identifies the main hydrothermal humification pathways of lignocellulosic waste, sets quantitative evaluation indicators to divide the working conditions of the main hydrothermal humification pathways, and establishes the quantitative relationship between the key intermediate product characteristics and the humic acid yield under each working condition; by constructing a process parameter database for hydrothermal preparation of humic acid from lignocellulosic waste via key reaction intermediates, a particle swarm algorithm optimized least squares support vector machine model (PSO-LS-SVM) algorithm is established for predicting the hydrothermal humification potential of lignocellulosic waste.

[0006] The present invention provides a method for predicting the hydrothermal humification potential of lignocellulosic waste. The method has the following specific steps: (1) hydrothermally carbonizing the lignocellulosic waste to prepare a reaction intermediate - hydrothermal charcoal, and then subjecting the hydrothermal charcoal to alkaline hydrothermal treatment to prepare humic acid; (2) using the average molecular formula unsaturation of the hydrothermal charcoal as a quantitative evaluation index, the hydrothermal conversion of the lignocellulosic waste into the humic acid is divided into two working conditions: insufficient hydrothermal carbonization and sufficient hydrothermal carbonization; (3) establishing a quantitative relationship between the average molecular formula unsaturation of the hydrothermal charcoal and the humic acid yield under each working condition; (4) constructing a process parameter database for hydrothermally preparing humic acid from the lignocellulosic waste via the hydrothermal charcoal; (5) using the hydrothermal conditions as input and the humic acid yield as output, establishing a particle swarm optimization least squares support vector machine model (PSO-LS-SVM) algorithm for predicting the hydrothermal humification potential of the lignocellulosic waste.

[0007] Furthermore, the lignocellulosic waste refers to biomass waste rich in hemicellulose, cellulose and lignin (referred to as the three elements) such as agricultural and forestry waste, gardening waste, fruit and vegetable waste, etc. The conditions for preparing hydrothermal charcoal by hydrothermal carbonization of the lignocellulosic waste raw materials are as follows: the hydrothermal temperature range is 160℃~250℃, the reaction time range is 1h~8h, the liquid-solid ratio is 2:1~20:1, and the hydrothermal pH is 0~7; subsequently, the conditions for preparing humic acid by alkaline hydrothermal treatment of the hydrothermal charcoal are: the hydrothermal temperature range is 160℃~250℃, the reaction time range is 1h~8h, the liquid-solid ratio is 2:1~20:1, and the hydrothermal pH is 12~14.

[0008] Furthermore, the average molecular formula unsaturation of the hydrothermal carbon is calculated by the carbon, hydrogen and nitrogen contents of the hydrothermal carbon: number of carbon atoms + 1 - (number of hydrogen atoms - number of nitrogen atoms) / 2.

[0009] Furthermore, sufficient hydrothermal carbonization means that the average molecular formula unsaturation of the hydrothermal char is ≤4, and the difference between the average molecular formula unsaturation of the hydrothermal char and the average molecular formula unsaturation of the lignocellulosic waste raw material is ≥1; insufficient hydrothermal carbonization means that the difference between the average molecular formula unsaturation of the hydrothermal char and the average molecular formula unsaturation of the lignocellulosic waste raw material is ≤1.

[0010] Furthermore, when the hydrothermal carbonization is sufficient, the average molecular formula unsaturation value (x) of the hydrothermal carbon is positively correlated with the humic acid yield (y) of the product, which can be calculated by the following formula: y=21.5x-16.8; when the hydrothermal carbonization is insufficient, the humic acid yield of the product is related to the amount of hemicellulose, cellulose and lignin contained in the hydrothermal carbon: y=the amount of hemicellulose contained in the hydrothermal carbon×0.5+the cellulose content×0.6+the lignin×61.6.

[0011] Furthermore, the process database for preparing humic acid from lignocellulosic waste via hydrothermal carbonization includes the characteristics of the lignocellulosic waste raw materials, the conditions of the hydrothermal carbonization, the characteristics of the hydrothermal charcoal, and the humic acid yield calculated based on the two conditions of sufficient and insufficient hydrothermal carbonization, wherein the characteristics of the lignocellulosic waste raw materials include dosage, three element content and carbon, hydrogen, nitrogen and oxygen content, and average molecular formula unsaturation; the conditions of the hydrothermal carbonization include reaction temperature, reaction time, liquid-solid ratio, hydrothermal pH, and container volume; the characteristics of the hydrothermal charcoal include carbon, hydrogen, nitrogen and oxygen content, average molecular formula unsaturation, total yield, ash content, and organic matter yield.

[0012] Furthermore, the PSO-LS-SVM model uses hydrothermal conditions as input items, specifically the amount of the lignocellulosic waste raw material, the three element content and the hydrothermal carbonization conditions (reaction temperature, reaction time, liquid-solid ratio, hydrothermal pH, container volume).

[0013] Furthermore, the PSO-LS-SVM model is established by writing code in Python, and the specific process is as follows:

[0014] The condition column inputs: the characteristics of the lignocellulosic waste raw materials and the conditions of the hydrothermal carbonization;

[0015] Result column output: mass of hydrothermal humic acid;

[0016] Data preprocessing: normalize the conditional data;

[0017] Data set division: The data is divided into a training set and a test set, wherein the training set accounts for 70% to 80% of the total data, and the test set accounts for 20% to 30% of the total data;

[0018] Modeling method: Use PSO-LS-SVM model to optimize the objective function R 2 To maximize, it is achieved by adjusting the particle swarm size (30-50), inertia weight (0.3-1), individual cognitive coefficient (1-2), and social experience coefficient (1-2);

[0019] Parameter optimization: Set the number of PSO iterations to 100 to 500, and search for the optimal combination of kernel function parameters γ and σ to make R 2 Optimization;

[0020] Performance Evaluation: Using R 2 , mean square error (MAE), mean square error (MSE) and root mean square error (RMSE) and other indicators to evaluate the performance of the vector machine model;

[0021] The vector machine model stability test: run 50 to 100 times, randomly set up the training set and the test set from the database, and obtain 50 to 100 groups of model prediction results.

[0022] The present invention has the following beneficial effects:

[0023] 1. The working conditions of hydrothermal conversion of lignocellulosic waste into humic acid involved in this method can provide data support for subsequent in-depth research and engineering applications;

[0024] 2. The PSO-LS-SVM prediction model constructed by this method has a wide range of applications and can evaluate many types of lignocellulosic wastes and complex hydrothermal conditions;

[0025] 3. The PSO-LS-SVM prediction model constructed by this method can save users' experimental costs and energy. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 Flowchart for the PSO-LS-SVM model to predict the hydrothermal humification potential of lignocellulosic waste;

[0027] Figure 2 The results of the PSO-LS-SVM model in Example 1 predicting the hydrothermal humification potential of lignocellulosic waste under the condition of sufficient hydrothermal carbonization are shown in Figure 1. (a) is the statistical result of the MAE value; (b) is the statistical result of the MSE value; (c) is the statistical result of the RMSE value; (d) is the R 2 Value statistics results;

[0028] Figure 3The results of the PSO-LS-SVM model in Example 2 predicting the hydrothermal humification potential of lignocellulosic waste when the hydrothermal carbonization is insufficient are shown in Figure 2. (a) shows the MAE value statistical result; (b) shows the MSE value statistical result; (c) shows the RMSE value statistical result; (d) shows the R 2 Value statistics results. DETAILED DESCRIPTION

[0029] All features disclosed in this specification, or steps in all methods or processes disclosed, except mutually exclusive features and / or steps, can be combined in any manner.

[0030] Any feature disclosed in this specification (including any additional claims and abstract), unless otherwise stated, may be replaced by other equivalent or alternative features having similar purposes. That is, unless otherwise stated, each feature is only an example of a series of equivalent or similar features.

[0031] Example 1

[0032] First, the program was set to input the characteristics of the lignocellulosic waste raw materials used for hydrothermal (including raw material mass, hemicellulose, cellulose and lignin content, and carbon, hydrogen and oxygen content) and hydrothermal carbonization conditions (including hydrothermal reaction temperature, reaction time, hydrothermal pH, liquid-solid ratio and container volume) as condition columns, and output the hydrothermal humic acid mass as the result column (see Tables 1-3). Secondly, according to the unsaturation of the raw materials and hydrothermal charcoal, the program was set to read the data in the data set that met the conditions for sufficient hydrothermal carbonization of lignocellulosic waste (Type 1). In order to unify the conditional variable domain, this study normalized all conditional data (i.e., the characteristics of lignocellulosic waste raw materials and hydrothermal carbonization conditions) between –1 and +1. Subsequently, 20% of the data that met Type 1 were extracted as the test set, and the remaining 80% of the data were used as the training set. The PSO-LS-SVM (the flow chart of the model is shown in the figure) constructed using the SVM module of the scikit-learn library and the PSO module of the Pyswarm library Figure 1 ), using the above condition column as input, predict the mass of hydrothermal humic acid derived from hydrothermal charcoal and compare it with the value stored in the result column. 2 To optimize the objective function, according to its numerical results (R 2The closer the value is to 1, the better the model fit is) and the particle swarm size, inertia weight, individual cognitive coefficient and social experience coefficient are continuously adjusted. The final parameters after optimization and adjustment are as follows: the particle swarm size is set to 40, the inertia weight, individual cognitive coefficient and social experience coefficient are set to 0.3, 1.5 and 1.5 respectively. The number of PSO iterations is set to 200 times, that is, the algorithm updates the position and velocity of each particle and searches 200 times in the particle swarm space to find the optimal kernel function parameter γ and parameter σ combination (γ ranges from 0.001 to 1, σ ranges from 0.01 to 10), so that R 2 Maximize. In addition, this study also uses mean square error (MAE), mean square error (MSE) and root mean square error (RMSE) to test the accuracy of the model. The entire program flow is set to run 50 times, that is, training sets and test sets are randomly set up from the constructed data set to obtain 50 sets of model prediction results, which are used to test the stability of the PSO-LS-SVM prediction model.

[0033] This example uses box plots to statistically analyze the evaluation index results of the PSO-LS-SVM model for predicting the hydrothermal humification potential of lignocellulosic waste, namely, the MAE value, MSE value, RMSE value, and R 2 When the hydrothermal carbonization of lignocellulosic waste is sufficient, the MAE of the 50 PSO-LS-SVM prediction models ranges from 0.56g to 1.80g, with a median of 1.15g; the MSE range is 1.06g. 2 ~11.46g 2 The median is 2.62g 2 ; RMSE ranged from 1.03g to 3.39g, with a median of 1.90g ( Figure 2 ). The smaller these values ​​are, the better the overall performance of the model. As the target optimization function of the model, R 2 The range was between 0.74 and 0.96, with a median of 0.89, indicating that the fit of these models was high.

[0034] Example 2

[0035] The program was set the same as in Example 1, and the data set read was the data that met the conditions of insufficient hydrothermal carbonization of lignocellulosic waste (Type 2). For the case of insufficient hydrothermal carbonization of lignocellulosic waste, the MAE of the 50 PSO-LS-SVM prediction models ranged from 0.19 g to 0.71 g, with a median of 0.34 g; the MSE ranged from 0.08 g to 0.71 g. 2 ~3.27g 2 The median is 0.55g 2 ; RMSE ranges from 0.28g to 1.81g, with a median of 0.74g ( Figure 3 ).2 The range is between 0.78 and 0.97, with a median of 0.94, indicating that the model can also predict the hydrothermal humification potential of lignocellulosic waste with high accuracy when hydrothermal carbonization is insufficient, and within the constructed data set, the used procedure can make the performance of the PSO-LS-SVM model stable.

[0036]

[0037]

[0038]

Claims

1. A method for predicting the hydrothermal humification potential of lignocellulosic waste, characterized in that: include: (1) firstly, the lignocellulosic waste is hydrothermally carbonized to prepare hydrothermal charcoal as a key reaction intermediate, and then the hydrothermal charcoal is subjected to alkaline hydrothermal treatment to prepare humic acid as a hydrothermal humification pathway; (2) using the average molecular formula unsaturation of the hydrothermal char as a quantitative evaluation index, dividing the working conditions of the hydrothermal conversion of the lignocellulosic waste into the humic acid into insufficient hydrothermal carbonization and sufficient hydrothermal carbonization; (3) establishing a quantitative relationship between the average molecular unsaturation of the hydrothermal carbon and the humic acid yield of the product under each operating condition; (4) constructing a process parameter database for preparing humic acid from the lignocellulosic waste by hydrothermal carbon alkali hydrothermal treatment; (5) Taking the hydrothermal conditions as input and the humic acid yield as output, a least squares support vector machine model optimized by a particle swarm algorithm is established to predict the yield of the hydrothermal humification of the lignocellulosic waste.

2. The method for predicting the hydrothermal humification potential of lignocellulosic waste according to claim 1, characterized in that: In step (1), the lignocellulosic waste refers to waste rich in hemicellulose, cellulose and lignin, such as agricultural and forestry waste, kitchen waste, greening waste, and biogas residue; the conditions for preparing the hydrothermal charcoal by acid hydrolysis of the lignocellulosic waste raw material are as follows: the hydrothermal temperature range is 160°C to 250°C, the reaction time range is 1h to 8h, the liquid-solid ratio is 2:1 to 20:1, and the hydrothermal pH is 0 to 7; the conditions for preparing humic acid by alkaline hydrothermal treatment of the hydrothermal charcoal are as follows: the hydrothermal temperature range is 160°C to 250°C, the reaction time range is 1h to 8h, the liquid-solid ratio is 2:1 to 20:1, and the hydrothermal pH is 12 to 14.

3. The method for predicting the hydrothermal humification potential of lignocellulosic waste according to claim 1, characterized in that: In step (2), the average molecular formula unsaturation of the hydrothermal carbon is calculated by the carbon, hydrogen and nitrogen contents of the hydrothermal carbon: number of carbon atoms + 1 - (number of hydrogen atoms - number of nitrogen atoms) / 2.

4. The method for predicting the hydrothermal humification potential of lignocellulosic waste according to claim 1, characterized in that: In step (2), sufficient hydrothermal carbonization means that the average molecular formula unsaturation of the hydrothermal carbon is ≤4, and the difference between the average molecular formula unsaturation of the hydrothermal carbon and the average molecular formula unsaturation of the lignocellulosic waste raw material is ≥1; insufficient hydrothermal carbonization means that the difference between the average molecular formula unsaturation of the hydrothermal carbon and the average molecular formula unsaturation of the lignocellulosic waste raw material is ≤1.

5. The method for predicting the hydrothermal humification potential of lignocellulosic waste according to claim 1, characterized in that: In step (3), when the hydrothermal carbonization is sufficient, the average molecular formula unsaturation value x of the hydrothermal carbon is positively correlated with the humic acid yield y of the product, which can be calculated by the following formula: y=21.5x-16.8; when the hydrothermal carbonization is insufficient, the humic acid yield of the product is related to the content of hemicellulose, cellulose and lignin contained in the hydrothermal carbon: y=the amount of hemicellulose contained in the hydrothermal carbon×0.5+the cellulose content×0.6+the lignin×61.

6.

6. The method for predicting the hydrothermal humification potential of lignocellulosic waste according to claim 1, characterized in that: In step (4), the process route database related to the hydrothermal preparation of humic acid from the lignocellulosic waste via the hydrothermal carbonization intermediate product specifically includes the characteristics of the lignocellulosic waste raw materials, the conditions of the hydrothermal carbonization, the characteristics of the hydrothermal carbon, and the humic acid yield calculated according to the two conditions of sufficient and insufficient hydrothermal carbonization, wherein the characteristics of the lignocellulosic waste raw materials include dosage, three element content and carbon, hydrogen, nitrogen and oxygen content, and average molecular formula unsaturation; the conditions of the hydrothermal carbonization include reaction temperature, reaction time, liquid-solid ratio, hydrothermal pH, and container volume; the characteristics of the hydrothermal carbon include carbon, hydrogen, nitrogen and oxygen content, average molecular formula unsaturation, total yield, ash content, and organic matter yield.

7. The method for predicting the hydrothermal humification potential of lignocellulosic waste according to claim 6, characterized in that: In step (5), the vector machine model uses hydrothermal conditions as input items, specifically the amount of the lignocellulosic waste raw material, the three element contents and the hydrothermal carbonization conditions.

8. The method for predicting the hydrothermal humification potential of lignocellulosic waste according to claim 6, characterized in that: In step (5), the vector machine model is established by writing code in Python, and the specific process is as follows: The condition column inputs: the characteristics of the lignocellulosic waste and the conditions of the hydrothermal carbonization; Result column output: mass of hydrothermal humic acid; Data preprocessing: normalize the conditional data; Data set division: The data is divided into a training set and a test set, wherein the training set accounts for 70% to 80% of the total data, and the test set accounts for 20% to 30% of the total data; Modeling method: Use the vector machine model to optimize the objective function R 2 To maximize, it is achieved by adjusting the particle swarm size (30-50), inertia weight (0.3-1), individual cognitive coefficient (1-2), and social experience coefficient (1-2); Parameter optimization: Set the number of iterations of the vector machine model to 100 to 500 times, and search for the optimal combination of kernel function parameters γ and parameters σ to make R 2 Optimization; Performance Evaluation: Using R 2 , mean square error, mean square deviation and root mean square error indicators to evaluate the performance of the vector machine model; The vector machine model stability test: run 50 to 100 times, randomly set up the training set and the test set from the database, and obtain 50 to 100 groups of model prediction results.

Citation Information

Patent Citations

  • Method for enhancing hydrothermal humification of lignocellulose waste through acid hydrothermal pretreatment

    CN113563603A

  • Method for hydrothermally preparing humic acid by using acid hydrothermal solution to circularly reinforce hydrothermal carbon and application of humic acid

    CN115109270A

  • Storage tank corrosion rate prediction method

    CN115510742A

  • Adsorbed water treatment scheme optimization method, device and equipment based on particle swarm

    CN116306321A

  • Method for hydrothermally converting municipal sludge into mineral-like humic acid

    CN117263476A