A method for predicting the final temperature of garden biomass carbonization

Abstract

The present invention relates to the field of biochar production from garden waste, and more particularly to a method for predicting the final carbonization temperature of garden biomass. The method comprises collecting data on factors influencing the final carbonization temperature of garden biomass and the final carbonization temperature; screening the data based on the carbonization yield; obtaining a training set; training the training set data to obtain a prediction model for the final carbonization temperature of garden biomass; testing the prediction model; and using the garden biomass carbonization final temperature prediction model for prediction. The optimal final carbonization temperature of garden biomass can be predicted based on key factors, thereby improving work efficiency while ensuring the production output of garden biomass carbonization, and increasing the scientific nature and controllability of the production process.

Description

Technical Field

[0001] The present invention relates to the field of production of garden waste biochar, and in particular to a method for predicting the final temperature of garden biomass carbonization. Background Art

[0002] Biomass energy is the fourth largest energy source after coal, oil, and natural gas, accounting for 14% of global primary energy. my country's biomass energy utilization rate is low, and energy consumption still primarily relies on fossil fuels. Therefore, efficient biomass energy utilization is of great significance. Garden biomass carbonization is a new type of clean, renewable biomass energy source. This technology can reduce the bioavailability of heavy metals in carbonized raw materials, kill pathogens and weeds at high temperatures, and reduce antibiotic content, thereby reducing the environmental risks of using carbonized raw materials. Furthermore, the prepared biochar retains a large amount of nutrients from the carbonized raw materials, which can improve soil fertility. High-temperature pyrolysis gives the biochar a large specific surface area and a rich pore structure, which offers significant advantages in water absorption and nutrient retention.

[0003] The biomass pyrolysis process is complex, mainly composed of cracking reactions and polycondensation reactions, with many intermediate reaction pathways. The pyrolysis reactions are mainly decarboxylation reactions, decarbonylation reactions, dehydration reactions, and reverse aldol condensation reactions, including the cracking of cellulose, hemicellulose, and lignin, the volatilization of light components in the cracking products, the decomposition and recombination of volatile products during precipitation, the polycondensation of cracking residues, further decomposition, and further polycondensation. The conditions of biomass pyrolysis, such as the type of raw materials, heating rate, final carbonization temperature, residence time, raw material moisture, particle size, etc., all affect the yield and composition of pyrolysis products to varying degrees. Among them, the final carbonization temperature has a significant impact on the type of pyrolysis products, the pyrolysis carbon yield, and the specific surface area of ​​biochar. In existing processes, it is impossible to reliably determine the appropriate final pyrolysis temperature based on the pyrolysis process parameters and the demand for target pyrolysis products. Summary of the Invention

[0004] The purpose of the present invention is to solve the above-mentioned related problems and design a method for predicting the final temperature of garden biomass carbonization. To achieve the above-mentioned purpose, the present invention provides the following scheme:

[0005] A method for predicting the final temperature of garden biomass carbonization comprises the following steps:

[0006] S1. Collect data on factors affecting the final carbonization temperature of garden biomass and the final carbonization temperature, and correlate the data on factors affecting the final carbonization temperature of garden biomass with the data on the final carbonization temperature;

[0007] S2. Screening the data on factors influencing the final carbonization temperature of forest biomass and the correlation between the final carbonization temperature and the final carbonization temperature based on the carbonization yield;

[0008] S3. Take 70%-80% of the filtered data as the training set and the remaining data as the test set, and number the data in the training set;

[0009] S4. The training set data is trained using a decision tree regression model to obtain a garden biomass carbonization final temperature prediction model;

[0010] S5. The garden biomass carbonization final temperature prediction model is tested using the test set data;

[0011] S6. Use the garden biomass carbonization final temperature prediction model to predict the required carbonization final temperature based on the data of the factors affecting the garden biomass carbonization final temperature.

[0012] As a further improvement of the present technical solution, the specific implementation of step S1 is as follows: determining the factors affecting the final temperature of garden biomass carbonization, including feed size, feed rate, feed moisture content, heating rate and pyrolysis time; collecting data on feed size, feed rate, feed moisture content, heating rate and pyrolysis time, and the final temperature of carbonization under the data conditions; the data set of factors affecting the final temperature of garden biomass carbonization ,in, is the feed size, is the feed rate, is the feed moisture content, is the heating rate, is the pyrolysis time; The final carbonization temperature corresponds to the data set of each group of influencing factors.

[0013] As a further improvement of the present technical solution, the specific implementation method of step S2 is: whether the carbonization yield obtained based on the influencing factors of the carbonization final temperature of the garden biomass and its corresponding carbonization final temperature reaches the preset standard is used as the basis for data screening, the carbonization yield that reaches the preset standard is qualified data and is retained, and the carbonization yield that does not reach the preset standard is unqualified data and is discarded; the carbonization yield is the mass ratio of the finally obtained garden biomass charcoal to the garden biomass raw material.

[0014] As a further improvement of this technical solution, the specific implementation of step S4 is as follows:

[0015] The training set after numbering is ; Select the optimal partition point , The corresponding data set of factors affecting the final temperature of garden biomass carbonization is ;

[0016] The optimal partition point is calculated as follows:

[0017]

[0018] Where, After division The corresponding training set data set, After division The corresponding training set data set; for All The average value of The output value of for All The average value of The output value of

[0019] right and Continue to divide at the optimal division point and calculate the output value until the stop condition is met. The final temperature prediction model of garden biomass carbonization is obtained as follows:

[0020]

[0021] Where M is the number of data sets that the training set is finally divided into; The output value for each data set;

[0022] .

[0023] As a further improvement of the present technical solution, the stopping condition is a preset number of divisions or a situation where the number of elements in a divided set is less than 2 after the division.

[0024] A system using a method for predicting the final temperature of garden biomass carbonization, characterized by comprising:

[0025] The data acquisition module is used to collect data on feed size, feed moisture content, feed rate, heating rate and pyrolysis time of the garden biomass carbonization system;

[0026] A control module for controlling the raw material pretreatment, feed rate, heating rate and pyrolysis time of the garden biomass carbonization system;

[0027] The machine learning and prediction module uses the garden biomass carbonization final temperature prediction method based on the data collected by the data acquisition module to output the carbonization final temperature;

[0028] The human-computer interaction module is used to set processing parameters and display data during the production process of the garden biomass carbonization system. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Various other advantages and benefits will become apparent to those skilled in the art by reading the detailed description of the preferred embodiment below. The accompanying drawings are only for the purpose of illustrating the preferred embodiment and are not to be considered as limiting the present invention;

[0030] Figure 1 This is a flow chart of a method for predicting the final temperature of garden biomass carbonization according to the present invention;

[0031] Figure 2 This is a schematic structural diagram of a system using a method for predicting the final temperature of garden biomass carbonization according to the present invention. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0033] Those skilled in the art will appreciate that, unless otherwise stated, the singular forms "a," "an," and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the stated features, integers, steps, operations, elements, and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0034] The present invention will be described in detail below with reference to the accompanying drawings.

[0035] Example 1

[0036] like Figure 1 As shown, a method for predicting the final temperature of garden biomass carbonization includes the following steps:

[0037] Data on factors affecting the final carbonization temperature of garden biomass and the final carbonization temperature were collected, and the data on factors affecting the final carbonization temperature of garden biomass were correlated with the final carbonization temperature data. The yield and composition of biomass pyrolysis products are affected by factors such as pyrolysis temperature, heating rate, pyrolysis atmosphere, and raw material pretreatment method. Different factors have different effects on biomass pyrolysis. When garden biomass carbonization equipment is used to produce garden biomass carbon, under the condition of a certain final carbonization product, the factors affecting the final carbonization temperature of garden biomass include feed size, feed rate, feed moisture content, heating rate, and pyrolysis time. The reason why feed size affects garden biomass carbonization is that if the particle size of garden waste is too large, it may cause the feed equipment to be blocked, affecting both the carbonization efficiency and the carbonization effect. In order to meet production requirements, the particle size of garden waste raw materials such as crushed branches and fallen leaves should be controlled at 3-5 cm. Feed moisture content affects the carbonization of garden biomass. If the feedstock is not effectively dried and its moisture content exceeds 20%, carbonization will produce a large amount of irritating gases. However, after drying, the moisture content of the feedstock must be below 20% to ensure sufficient and effective carbonization after entering the carbonization furnace. The heating rate also affects the carbonization of garden biomass. A higher heating rate promotes the release of volatiles, making the pyrolysis reaction more likely to proceed. A higher heating rate increases the porosity of the biochar. Slower heating rates make it easier to obtain biochar from pyrolyzed biomass. Faster heating rates shorten the residence time of the biomass in each temperature range, reducing the likelihood of secondary reactions of the volatiles at high temperatures and improving the yields of gaseous and liquid products. The reason why pyrolysis time affects the carbonization of garden biomass is that under the conditions of constant pyrolysis temperature and heating rate, the extension of reaction residence time will increase the yield of biochar and also have a certain impact on the ash content and elemental composition of biochar. Shortening the residence time of pyrolysis gas in the reactor helps the pyrolysis gas products to separate from the particle surface, reduce secondary reactions, and improve the yield and quality of bio-oil. Data on feed size, feed rate, feed moisture content, heating rate, pyrolysis time and final carbonization temperature are collected to form a data set of factors affecting the final temperature of garden biomass carbonization. ,in, is the feed size, is the feed rate, is the feed moisture content, is the heating rate, is the pyrolysis time; is the final carbonization temperature corresponding to each group of influencing factors. , ) constitute an instance.

[0038] The data after the influencing factors of the carbonization final temperature of garden biomass and the correlation between the carbonization final temperature and the carbonization final temperature were screened according to the carbonization yield. The garden biomass carbonization equipment was used to produce garden biomass carbon, and the carbonization yield of the obtained carbonized product was 20.50±1.98%. Therefore, according to the ratio of 18.52%, ( , ) were screened and the ones with carbonization rate of not less than 18.52% were retained ( , ), for less than 18.52% ( , ) were eliminated according to abnormal data. The carbonization yield is the ratio of the mass of biomass carbon obtained to the mass of the garden biomass fed.

[0039] Take 70%~80% of the filtered data as the training set, and the remaining data as the test set, and number the data in the training set. , ) as qualified data, use subscripts for all ( , The numbered data is divided into two parts, 70% to 80% of the data is used as the training set, and the remaining 20% ​​to 30% of the data is used as the test set to test the training results.

[0040] The training set data is trained using a decision tree regression model to obtain a garden biomass carbonization final temperature prediction model.

[0041] The numbered training set is Set the condition for the decision tree to stop growing to 3 times. That is, the decision tree regression model will stop growing after 4 times.

[0042] Select the optimal partition point , The corresponding data set of factors affecting the final temperature of garden biomass carbonization is ; The optimal partition point is calculated as follows:

[0043]

[0044] Where, After division The corresponding training set data set, After division The corresponding training set data set; for All The average value of The output value of for All The average value of The output value of . Substitute i=1~n into the traversal calculation to determine the optimal partition point, which is a split growth.

[0045] right and Then select the optimal division point and continue the division. It will be split into two data sets, and , and get the output values ​​respectively and , It will also split into two data sets and get the output value. This is the second split growth.

[0046] right Then select the optimal division point to continue division. It will be split into two data sets, and , and get the output values ​​respectively and The other sets are further split and the corresponding output values ​​are obtained. Finally, after 3 splits, the decision tree stops growing and 8 data sets and 8 output values ​​are obtained.

[0047] Therefore, the prediction model of the final temperature of garden biomass carbonization after three splits is:

[0048]

[0049] Where, The output value for each data set; Calculate as follows:

[0050]

[0051] That is, each time the split occurs, the set of optimal partition points is Take 1, the corresponding Participate in the calculation of the prediction model; the set where the optimal partition point is not Take 0, the corresponding Does not participate in the calculation of the prediction model.

[0052] The garden biomass carbonization final temperature prediction model is tested using the test set data. After the garden biomass carbonization final temperature prediction model is established, the model is tested using the test set consisting of the remaining data to determine the accuracy of the model and whether there are any problems such as overfitting.

[0053] The garden biomass carbonization final temperature prediction model is used to predict the required carbonization final temperature based on the data of the factors affecting the garden biomass carbonization final temperature. When using garden biomass carbonization equipment for production, the feed size, feed rate, feed moisture content, heating rate, and pyrolysis time data are set in advance according to the production needs. The garden biomass carbonization final temperature prediction model determines the appropriate carbonization final temperature based on the production parameters and automatically controls the equipment according to this temperature.

[0054] Example 2

[0055] like Figure 2 As shown, a system using a method for predicting the final temperature of garden biomass carbonization includes:

[0056] The data acquisition module is used to collect data on feed size, feed moisture content, feed rate, heating rate, pyrolysis time and carbonization final temperature of the garden biomass carbonization system;

[0057] A control module is used to control the raw material pretreatment, feed rate, heating temperature rise rate, pyrolysis time and carbonization final temperature of the garden biomass carbonization system;

[0058] The machine learning and prediction module uses the garden biomass carbonization final temperature prediction method based on the data collected by the data acquisition module to output the carbonization final temperature;

[0059] The human-computer interaction module is used to set processing parameters and display data during the production process of the garden biomass carbonization system.

[0060] Before the start of processing, the processing parameters are input into the human-computer interaction module according to the needs. The processing parameters include feed size, feed rate, feed moisture content, heating rate, and pyrolysis time. Based on the processing parameters, the machine learning and prediction module uses the trained garden biomass carbonization final temperature prediction model to predict the optimal carbonization final temperature under the parameters. The control module controls the garden biomass carbonization production process according to the input processing parameters and the predicted carbonization final temperature. During the production process, the various processing parameters and carbonization final temperature data are collected in real time by the data acquisition module and fed back to the human-computer interaction module to facilitate the monitoring of the production process.

[0061] During the production process, the processing parameters can be modified at any time through the human-computer interaction module as needed. After the processing parameters are modified, the machine learning and prediction module obtains the new optimal carbonization final temperature, and adjusts the processing parameters and carbonization final temperature through the control module to achieve fully automatic control of garden biomass carbonization production.

[0062] In summary, the method for predicting the final temperature of garden biomass carbonization proposed in the present invention can predict the optimal final temperature of garden biomass carbonization based on key factors such as feed size, feed rate, feed moisture content, heating rate, and pyrolysis time. On the basis of ensuring the production output of garden biomass carbonization, it improves work efficiency and increases the scientificity and controllability of the production process.

Claims

1. A garden biomass carbonization final temperature prediction system, characterized in that: include: The data acquisition module is used to collect data on feed size, feed moisture content, feed rate, heating rate, pyrolysis time and carbonization final temperature of the garden biomass carbonization system; A control module is used to control the raw material pretreatment, feed rate, heating temperature rise rate, pyrolysis time and carbonization final temperature of the garden biomass carbonization system; The machine learning and prediction module associates the data of the factors influencing the final carbonization temperature of the garden biomass with the data of the final carbonization temperature based on the data of the factors influencing the final carbonization temperature of the garden biomass and the final carbonization temperature collected by the data collection module; screens the data after the factors influencing the final carbonization temperature of the garden biomass are associated with the final carbonization temperature based on the carbonization yield; takes 70% to 80% of the screened data as a training set, and the remaining data as a test set, and numbers the data in the training set; The training set data is trained using a decision tree regression model to obtain a garden biomass carbonization final temperature prediction model; the garden biomass carbonization final temperature prediction model is tested using a test set data; Use the garden biomass carbonization final temperature prediction model to predict the required carbonization final temperature based on the data of the factors affecting the garden biomass carbonization final temperature; The machine learning and prediction module uses the carbonization final temperature, an influencing factor of the garden biomass carbonization final temperature, and the carbonization yield obtained by the corresponding carbonization final temperature as a data screening basis. The carbonization yield that meets the preset standard is qualified data and retained, and the carbonization yield that does not meet the preset standard is unqualified data and discarded; the carbonization yield is the mass ratio of the finally obtained garden biomass charcoal to the garden biomass raw material; The human-computer interaction module is used to set processing parameters and display data during the production process of the garden biomass carbonization system.

2. A garden biomass carbonization final temperature prediction system according to claim 1, characterized in that: The machine learning and prediction module determines that the factors affecting the final carbonization temperature of garden biomass include feed size, feed rate, feed moisture content, heating rate and pyrolysis time; collects data on feed size, feed rate, feed moisture content, heating rate and pyrolysis time and the final carbonization temperature under the data conditions; and collects data on the factors affecting the final carbonization temperature of garden biomass. Among them, x1 is the feed size, x2 is the feed rate, x3 is the feed moisture content, x4 is the heating rate, and x5 is the pyrolysis time; y i is the final carbonization temperature corresponding to the data set of each influencing factor.

3. A garden biomass carbonization final temperature prediction system according to claim 1, characterized in that: The machine learning and prediction module uses the final carbonization temperature, an influencing factor of the final carbonization temperature of the garden biomass, and the carbonization yield obtained by the corresponding final carbonization temperature as a basis for data screening. The data with a carbonization yield that meets the preset standard are qualified and retained, and the data with a carbonization yield that does not meet the preset standard are unqualified and discarded; the carbonization yield is the mass ratio of the finally obtained garden biomass charcoal to the garden biomass raw material.

4. A garden biomass carbonization final temperature prediction system according to claim 1, characterized in that: The training set after the machine learning and prediction module is numbered is Select the optimal partition point y j =s,y j The corresponding data set of factors affecting the final temperature of garden biomass carbonization is The optimal partition point is calculated as follows: In the formula, R1(j,s) is the y after division i ≤s corresponds to the data set of the training set, R2(j,s) is the data set of y after division i > The data set of the training set corresponding to s; c1 is all y in R1(j,s) i The average value of is the output value of R1(j,s); c2 is the average value of all y in R2(j,s). i The average value is the output value of R2(j,s); R1(j,s) and R2(j,s) are divided at the optimal division point and the output value is calculated until the stop condition is met. The prediction model for the final temperature of garden biomass carbonization is obtained as follows: Where M is the number of data sets that the training set is finally divided into; c i The output value for each data set; 5. A garden biomass carbonization final temperature prediction system according to claim 4, characterized in that: The stopping condition is a preset number of divisions or a situation where the number of elements in a divided set is less than 2 after division.

Images