A method, device, medium and product for predicting co-pyrolysis products of biomass and plastics

Abstract

The present application discloses a method, device, medium and product for predicting the co-pyrolysis products of biomass and plastics, and relates to the field of agricultural waste treatment. The method comprises obtaining the type of biomass to be decomposed and the composition ratio of plastics; according to the type of biomass to be decomposed and the composition ratio of plastics, using a co-pyrolysis product prediction model to determine the co-pyrolysis product ratio; wherein the co-pyrolysis product prediction model is obtained by training a deep neural network model using a training data set; the training data set is determined based on the type of training biomass and the composition ratio of training plastics using high-throughput experimental simulation; the training data set includes the type of training biomass, the composition ratio of training plastics and the corresponding co-pyrolysis product ratio. The present application improves the accuracy of the prediction of the co-pyrolysis products of biomass and plastics.

Description

Technical Field

[0001] The present application relates to the field of agricultural waste treatment, and in particular to a method, device, medium and product for predicting the co-pyrolysis products of biomass and plastic. Background Art

[0002] The higher plant biomass raw materials in agricultural waste have large output and broad application prospects. If combined with efficient and pollution-free biomass pyrolysis (irreversible thermochemical reaction under anaerobic or anoxic conditions) technology, it can produce abundant clean energy such as biogas and bio-oil and multifunctional biochar materials, thereby reducing dependence on fossil energy.

[0003] Plant species are diverse in number and variety. The number of currently identifiable species alone has reached more than 400,000, and their structures are complex and heterogeneous. The diverse internal characteristics and complex component structure of biomass make it difficult to accurately control and predict its processing technology and products. It also affects the properties and potential applications of the final products, limiting its breadth of application. Although the components of different types of biomass are quite different, cellulose, hemicellulose and lignocellulose, as their main components, constitute the main components of plant cell walls, accounting for more than 90% of dry plant cells. Determining the distribution of cellulose, hemicellulose and lignocellulose in plants is very important for understanding the cross-linked chemical structure of biomass and fully realizing the potential of biomass in various applications. Therefore, in-depth research on the composition changes and structural transformations of biomass cells during the actual pyrolysis process, in-depth exploration of the decomposition behavior of each organic component of biomass and its influence on the pyrolysis products, is of great significance for clarifying the mechanism of the pyrolysis process of biomass.

[0004] The landfill method is to send waste plastics to landfills as garbage. It is a traditional method widely used to deal with waste plastics and is still the main method used in countries around the world. However, this simple and direct physical method only treats plastics as garbage and does not reuse them, which is a huge waste of resources. In addition, it is almost impossible for plastics to be completely decomposed in nature. The use of landfills seriously hinders groundwater infiltration, causing secondary pollution to the land, increasing the pressure on the use of land resources, and thus causing greater hidden dangers and hazards.

[0005] The co-processing and utilization of agricultural waste biomass and plastics will play an important role in China's sustainable development and environmental protection. Studies have found that the co-pyrolysis of biomass and plastics mainly produces three-state products, namely liquid bio-oil, gaseous mixed gas and solid biochar. Among them, solid biochar can be processed into activated carbon, and it will show different adsorption properties for different substances due to its different morphological structures. The bio-oil produced by co-pyrolysis contains more aliphatic hydrocarbons, esters and alcohols than the bio-oil produced by pure biomass pyrolysis, and the oxygen-containing and nitrogen-containing products are reduced while the hydrocarbon ratio is increased. Since the content of bio-oil hydrocarbons obtained by co-pyrolysis is increased, it is more similar to the components of fuel, and thus has a broader application prospect. It is worth noting that adding plastics to biomass usually reduces its activation energy, thereby producing a synergistic effect and increasing the yield and quality of liquid products. Due to the high carbon content of plastics, the hydrocarbons with higher calorific value in the co-pyrolysis gas products are increased compared with pure biomass, which increases the utilization value of biogas. In addition, co-pyrolysis can also inhibit the production of carbon dioxide and promote the production of methane, achieving the purpose of reducing carbon dioxide emissions while increasing the source of natural gas.

[0006] However, the composition of the raw materials has an important influence on the ratio of pyrolysis products. Previous studies have evaluated how the presence of plastics affects the characteristics of the produced biochar. The study found that the biochar yield of the pyrolysis process is mainly determined by the biomass raw material. A low proportion of plastics has a positive effect on the charcoal yield, while a high proportion of plastics has a negative effect on the charcoal yield. The thermochemical properties of cellulose, hemicellulose, lignin, etc. in biomass are different, and the chemical reactions between the components during the co-pyrolysis of biomass and plastics are intricate. Existing technical research has not yet clearly revealed the material transformation pathways, element migration paths, and chemical reactions between gaseous substances during the co-pyrolysis of biomass and plastics, resulting in the inability to carry out large-scale industrial applications of pyrolysis. The mechanism of the co-pyrolysis reaction of biomass and plastics, especially the cross-reactions between the components, such as the chemical reactions between gaseous substances, and the reactions between gaseous substances and high-temperature biochar, still needs to be explored. Summary of the invention

[0007] The purpose of this application is to provide a method, device, medium and product for predicting the co-pyrolysis products of biomass and plastics, so as to improve the accuracy of the prediction of the co-pyrolysis products of biomass and plastics.

[0008] To achieve the above objectives, this application provides the following solutions:

[0009] In a first aspect, the present application provides a method for predicting co-pyrolysis products of biomass and plastics, comprising:

[0010] Obtain the types of biomass to be decomposed and the proportion of plastic composition;

[0011] According to the types of biomass to be decomposed and the composition ratio of plastics, the co-pyrolysis product prediction model is used to determine the co-pyrolysis product ratio; wherein the co-pyrolysis product prediction model is obtained by training a deep neural network model using a training data set; the training data set is based on the types of training biomass and the composition ratio of training plastics, and is determined by high-throughput experimental simulation; the training data set includes the types of training biomass, the composition ratio of training plastics and the corresponding co-pyrolysis product ratios.

[0012] Optionally, according to the type of biomass to be decomposed and the composition ratio of the plastic, a co-pyrolysis product prediction model is used to determine the co-pyrolysis product ratio, specifically including:

[0013] Standardizing the types of biomass to be decomposed and the composition ratio of plastics to obtain the types of biomass after treatment and the composition ratio of plastics after treatment;

[0014] According to the type of biomass after treatment and the composition ratio of plastic after treatment, the ratio of co-pyrolysis products is determined by using the co-pyrolysis product prediction model.

[0015] Optionally, based on the types of training biomass and the composition ratio of training plastics, a training data set is determined by high-throughput experimental simulation, specifically including:

[0016] Construct the co-pyrolysis reaction mechanism of biomass and plastics; the co-pyrolysis reaction mechanism of biomass and plastics is the influence of biomass type and plastic composition ratio on co-pyrolysis products;

[0017] According to the co-pyrolysis reaction mechanism of biomass and plastics, a kinetic model of the co-pyrolysis reaction of biomass and plastics was constructed;

[0018] Based on the training biomass types, training plastic composition ratios, and the kinetic model of the co-pyrolysis reaction of biomass and plastics, the SuperPro Designer software was used to simulate the reaction process, and the component object model was used to drive the high-throughput experimental simulation to determine the co-pyrolysis products and the co-pyrolysis product ratios, thereby constructing a high-throughput biomass pyrolysis data set;

[0019] Preprocessing the high-throughput biomass pyrolysis data set to obtain a processed high-throughput biomass pyrolysis data set;

[0020] The processed high-throughput biomass pyrolysis dataset is divided into training dataset, validation dataset and test dataset.

[0021] Optionally, it also includes:

[0022] Based on a high-throughput biomass pyrolysis dataset, a waste reduction algorithm is used to determine the environmental impact of co-pyrolysis of biomass and plastics; the environmental impact of co-pyrolysis of biomass and plastics includes carbon dioxide emissions.

[0023] Optionally, it also includes:

[0024] According to the type of biomass to be decomposed and the proportion of plastic composition, the co-pyrolysis product prediction model is used to determine the impact of co-pyrolysis of biomass and plastic on the environment.

[0025] Optionally, the deep neural network model is trained using the training data set, specifically including:

[0026] Taking the types of processed training biomass and the composition ratio of processed training plastics in the current training data set as input and the corresponding co-pyrolysis product ratios in the current training data set as output, the current deep neural network model is trained to obtain a trained deep neural network model;

[0027] The trained deep neural network model is verified and tested using the verification data set and the test data set respectively to obtain the initial deep neural network model;

[0028] Obtain the types of biomass samples and the composition ratio of plastic samples, use the current initial deep neural network model to determine the ratio of co-pyrolysis products of biomass samples and plastic samples, and draw a piper three-line diagram;

[0029] Determine whether the current initial deep neural network model reaches the preset prediction accuracy;

[0030] If yes, the current initial deep neural network model is used as the co-pyrolysis product prediction model;

[0031] If not, based on the types of biomass samples, the composition ratio of plastic samples, and the kinetic model of the co-pyrolysis reaction of biomass and plastics, the SuperPro Designer software is used to simulate the reaction process, and the component object model is used to drive the high-throughput experimental simulation to determine the co-pyrolysis products and the ratio of the co-pyrolysis products, and the current training data set is expanded to obtain the expanded training data set;

[0032] The expanded training data set is used as the current data set, and the result is "using the types of processed training biomass and the composition ratio of processed training plastics in the current training data set as input, and the corresponding co-pyrolysis product ratios in the current training data set as output to train the current deep neural network model and obtain a trained deep neural network model".

[0033] In a second aspect, the present application provides a computer device comprising: 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 any of the above-described methods for predicting co-pyrolysis products of biomass and plastic.

[0034] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described methods for predicting co-pyrolysis products of biomass and plastic.

[0035] In a fourth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements any of the above-mentioned methods for predicting the co-pyrolysis products of biomass and plastic.

[0036] According to the specific embodiments provided in this application, this application has the following technical effects:

[0037] The present application provides a method, device, medium and product for predicting the co-pyrolysis products of biomass and plastics, obtaining the types of biomass to be decomposed and the composition ratio of plastics; according to the types of biomass to be decomposed and the composition ratio of plastics, using the co-pyrolysis product prediction model, determine the co-pyrolysis product ratio; wherein the co-pyrolysis product prediction model is obtained by training a deep neural network model using a training data set; the training data set is determined based on the types of training biomass and the composition ratio of training plastics using high-throughput experimental simulation; the training data set includes the types of training biomass, the composition ratio of training plastics and the corresponding co-pyrolysis product ratio. The present application improves the accuracy of the prediction of the co-pyrolysis products of biomass and plastics. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 A schematic flow chart of a method for predicting co-pyrolysis products of biomass and plastics provided in one embodiment of the present application;

[0040] Figure 2 A flowchart of the biomass and plastic co-pyrolysis product prediction method in practical application of the present application;

[0041] Figure 3 Construct a flow chart for the co-pyrolysis product prediction model;

[0042] Figure 4 A piper three-line diagram provided for an embodiment of the present application;

[0043] Figure 5 A schematic diagram of the electronic device structure provided for this application;

[0044] Figure 6Schematic diagram of the structure of the biomass and plastic co-pyrolysis product prediction device provided in this application;

[0045] Figure 7 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0047] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0048] In an exemplary embodiment, Figure 1 and Figure 2 As shown, a method for predicting co-pyrolysis products of biomass and plastics is provided, comprising the following steps:

[0049] S1: Obtain the type of biomass to be decomposed and the proportion of plastic composition.

[0050] S2: According to the types of biomass to be decomposed and the composition ratio of plastics, the co-pyrolysis product prediction model is used to determine the co-pyrolysis product ratio; wherein the co-pyrolysis product prediction model is obtained by training a deep neural network model using a training data set; the training data set is based on the types of training biomass and the composition ratio of training plastics, and is determined by high-throughput experimental simulation; the training data set includes the types of training biomass, the composition ratio of training plastics and the corresponding co-pyrolysis product ratios.

[0051] As an optional implementation method, according to the type of biomass to be decomposed and the composition ratio of plastics, a co-pyrolysis product prediction model is used to determine the co-pyrolysis product ratio, specifically including:

[0052] The types of biomass to be decomposed and the proportion of plastic composition are standardized to obtain the types of biomass after treatment and the proportion of plastic composition after treatment.

[0053] According to the type of biomass after treatment and the composition ratio of plastic after treatment, the ratio of co-pyrolysis products is determined by using the co-pyrolysis product prediction model.

[0054] As an optional implementation, it also includes:

[0055] According to the type of biomass to be decomposed and the proportion of plastic composition, the co-pyrolysis product prediction model is used to determine the impact of co-pyrolysis of biomass and plastic on the environment.

[0056] As an optional implementation, based on the types of training biomass and the composition ratio of training plastics, a training data set is determined by high-throughput experimental simulation, specifically including:

[0057] Step 1: Construct the co-pyrolysis reaction mechanism of biomass and plastic; the co-pyrolysis reaction mechanism of biomass and plastic is the influence of biomass type and plastic composition ratio on the co-pyrolysis products.

[0058] In practical applications, the co-pyrolysis reaction mechanism of biomass and plastics is constructed to reveal the effects of the contents of cellulose, hemicellulose, and lignocellulose in biomass and the contents of polypropylene (PP), low-density polyethylene (LDPE), and polystyrene (PS) in plastics on the co-pyrolysis products.

[0059] The biomass pyrolysis reaction mechanism is proposed by combining reaction kinetics and experimental data to explore the influence of cellulose, hemicellulose and lignocellulose content on the biomass pyrolysis reaction path.

[0060] The cellulose, hemicellulose, and lignocellulose contents were obtained by staining the cross-section of the biomass with safranin fast green and observing it under a microscope.

[0061] The role of reaction kinetics is to calculate the reaction rate and thus understand the reaction mechanism. Biomass is mainly composed of cellulose, hemicellulose, and lignocellulose. Studying the pyrolysis reaction kinetics of cellulose, hemicellulose, and lignocellulose can help analyze the reaction degree of different components and thus predict the reaction products.

[0062] The reaction kinetics will be studied using two approaches: model-based kinetics and model-free kinetics.

[0063] The model-based dynamics are represented using Arrhenius functions:

[0064]

[0065] Where k is the reaction rate constant, A is the prefactor, E is the activation energy, R is the gas constant, and T is the temperature.

[0066] As an optimization technique, model-free methods are applied to improve the accuracy of biomass pyrolysis kinetics. One commonly used method is to use the Starink equation by integrating the Flynn-Wall-Ozawa (FWO) and Kissinger-Akira-Sunose (KAS) isoconversion methods:

[0067] The formula for calculating activation energy is as follows, because the activation energy data is related to the rate constant, and the rate constant determines the subsequent products.

[0068]

[0069] In the formula, the activation energy E a The slope of ln(β / T^1.8) versus 1 / T is obtained, where β is the thermal decomposition rate, T is the velocity, and C s is a constant.

[0070] By using a pyrolysis gas analysis mass spectrometer Py-GC / MS (Pyrolyzer-Gas Chromatography-Mass Spectrometry) coupled to a gas chromatography / mass spectrometer, the pyrolysis products were separated by a GC column and qualitatively analyzed by a mass spectrometry system to obtain information about the composition and microstructure, and then to infer the pyrolysis reaction path of the compound; the peak area normalization method was used to compare the masses of all components to estimate the relative content of the pyrolysis products; a small amount of inorganic elements was quantitatively analyzed by inductively coupled plasma optical emission spectrometry ICP-OES (Inductively Coupled Plasma Optical Emission Spectrometry), and solid samples were analyzed by acid digestion method.

[0071] In step 1, the co-pyrolysis mechanism of biomass and plastics is constructed by co-pyrolyzing biomass and plastics, analyzing the product characteristics, and revealing the prediction of the co-pyrolysis products of plastics and different types of biomass.

[0072] The plastics are three common waste plastics, including polypropylene PP, low-density polyethylene LDPE and polystyrene PS, with contents ranging from 0 to 100% respectively.

[0073] Specifically, a biomass was selected for co-pyrolysis with three plastics, and the co-pyrolysis products were collected by changing the ratio of the three plastics, and the yields of gas, liquid and solid products were calculated. The gas composition was analyzed by gas chromatograph (GC), and the light liquid component was analyzed by gas chromatography-mass spectrometer (GC-MS), so as to characterize the characteristics of the co-pyrolysis products. Among them, the product characteristics include the composition of gas products, the composition of liquid products, etc.

[0074] Step 2: Based on the co-pyrolysis reaction mechanism of biomass and plastics, a kinetic model of the co-pyrolysis reaction of biomass and plastics is constructed.

[0075] In practical applications, a kinetic model of the co-pyrolysis reaction of biomass and plastics is built based on the co-pyrolysis mechanism of biomass and plastics in step 1. Based on the kinetic model of the co-pyrolysis reaction of biomass and plastics, a high-throughput experimental simulation framework is developed to achieve accurate prediction of the co-pyrolysis reaction of biomass and plastics. It can not only predict the yield and quality of solid, gas and liquid products, but also predict the impact on the environment during the co-pyrolysis reaction, such as the emission of carbon dioxide gas.

[0076] The high-throughput experimental simulation framework determines key reaction simulation parameters such as reaction rate based on the experiments and results of step 1 to form a kinetic model for the co-pyrolysis reaction of biomass and plastics.

[0077] Step 3: Based on the training biomass types, training plastic composition ratios, and the co-pyrolysis reaction kinetic model of biomass and plastics, use SuperPro Designer software to simulate the reaction process, use the component object model to drive high-throughput experimental simulation, determine the co-pyrolysis products and the co-pyrolysis product ratios, and thus construct a high-throughput biomass pyrolysis dataset.

[0078] In practical applications, based on the kinetic model of co-pyrolysis of biomass and plastics, the reaction process was simulated using SPD (SuperProDesigner, a comprehensive process simulation and optimization software platform) software. The Component ObjectModule (COM) of the process model was used to drive high-throughput process simulation in Excel and Python computing environments to construct a high-throughput biomass pyrolysis dataset.

[0079] During the high-throughput experimental process simulation, the "user-defined" function allows various existing materials (different biomass and plastics) to be added to the model, thereby expanding the flexibility and applicability of the model and allowing the framework to achieve high efficiency while ensuring the accuracy of the model.

[0080] As an optional implementation, it also includes:

[0081] Based on a high-throughput biomass pyrolysis dataset, a waste reduction algorithm is used to determine the environmental impact of co-pyrolysis of biomass and plastics; the environmental impact of co-pyrolysis of biomass and plastics includes carbon dioxide emissions.

[0082] In practical applications, the assessment of environmental impact is based on the constructed high-throughput biomass pyrolysis dataset, and the Waste Reduction (WAR) algorithm is used to high-throughput evaluate the energy consumption and environmental impact of the co-pyrolysis process.

[0083] Step 4: Preprocess the high-throughput biomass pyrolysis dataset to obtain a processed high-throughput biomass pyrolysis dataset.

[0084] Step 5: Divide the processed high-throughput biomass pyrolysis dataset into a training dataset, a validation dataset, and a test dataset.

[0085] As an optional implementation method, the deep neural network model is trained using the training data set, specifically including:

[0086] The current deep neural network model is trained with the types of processed training biomass and the composition ratio of processed training plastics in the current training data set as input and the corresponding co-pyrolysis product ratios in the current training data set as output to obtain a trained deep neural network model.

[0087] The trained deep neural network model is verified and tested using the validation data set and the test data set respectively to obtain the initial deep neural network model.

[0088] Obtain the types of biomass samples and the composition ratio of plastic samples, use the current initial deep neural network model to determine the ratio of co-pyrolysis products of biomass samples and plastic samples, and draw a piper three-line diagram.

[0089] Determine whether the current initial deep neural network model reaches the preset prediction accuracy.

[0090] If so, the current initial deep neural network model is used as the co-pyrolysis product prediction model.

[0091] If not, based on the types of biomass samples, the composition ratio of plastic samples and the kinetic model of co-pyrolysis reaction of biomass and plastic, the SuperPro Designer software is used to simulate the reaction process, and the component object model is used to drive the high-throughput experimental simulation to determine the co-pyrolysis products and the ratio of co-pyrolysis products, and the current training data set is expanded to obtain the expanded training data set.

[0092] The expanded training data set is used as the current data set, and the result is "using the types of processed training biomass and the composition ratio of processed training plastics in the current training data set as input, and the corresponding co-pyrolysis product ratios in the current training data set as output to train the current deep neural network model and obtain a trained deep neural network model".

[0093] In practical applications, Python is used to build an active deep learning environment, and the tensorflow library is used to build a deep neural network model. According to the simulation experimental results (including product components, product yields, and environmental impacts), the product distribution can be accurately classified. By classifying the simulation results, the yields of oil, gas, and carbon in the products can be obtained, and other product quality information can be obtained, such as H in the gas product. 2 At the same time, this classification method can also reveal whether the co-pyrolysis process is environmentally friendly. Through these analyses, the inherent laws between biomass reactants and plastic additives and product distribution are revealed.

[0094] like Figure 3 As shown in the figure, the construction process of the co-pyrolysis product prediction model is divided into four stages: "training", "exploration", "expansion" and "labeling".

[0095] In the "training" phase, the 1,000 data points generated by the high-throughput experimental simulation framework are preprocessed, including data cleaning, standardization, category labeling, and splitting into training sets, validation sets, and test sets. Based on the active learning method, a deep neural network model will be built during the iteration process, and the processed high-throughput biomass pyrolysis dataset will be imported to train and obtain the initial deep neural network model. While training, accuracy verification is carried out, and the model architecture and hyperparameters are adjusted according to the accuracy of the validation dataset. The F1 score and AUC curve are used to evaluate the model accuracy, and the model is verified using the test dataset.

[0096] In the "exploration" stage, the biomass and plastic sample components are used as input, and the trained deep learning model is used to obtain the distribution of data points in the piper three-line diagram.

[0097] The “expand” phase uses the biomass and plastic sample components used in the “exploration” phase as input to calculate SPD and WAR.

[0098] The "marking" stage will determine whether the deep neural network model has made a successful prediction and map the results to the piper three-line diagram to expand the training data set. Ultimately, it will be based on the iteration of the four stages of "training", "exploration", "expansion" and "marking" until a sufficient training data set is obtained to generate an accurate deep neural network model.

[0099] Combined with the piper three-line diagram and active learning algorithm, the predicted products and environmental assessment data (including human ingestion toxicity potential (HTPI), human exposure toxicity potential (HTPE), terrestrial toxicity potential (TTP), aquatic toxicity potential (ATP), global warming potential (GWP), ozone depletion potential (ODP), photochemical oxidation potential (PCOP) and acidification potential (AP)) are accurately classified and analyzed, and the impact of the co-pyrolysis process on the environment is evaluated based on the environmental assessment data. The results of co-pyrolysis of different biomass and plastics are divided into (1) mainly gas production-environmentally friendly, (2) mainly gas production-environmentally unfriendly, (3) mainly oil production-environmentally friendly, etc. The reactants can also be reversed according to the products and environmental requirements. After each reverse, the algorithm will re-calculate and verify the model based on the reversed results, thus forming an automatically growing biomass-plastic co-pyrolysis platform.

[0100] The self-growing biomass-plastic pyrolysis platform (co-pyrolysis product prediction model) is an automatic and systematically updated and iterated high-throughput biomass pyrolysis active learning framework, which is formed by coupling with experiments and continuously updating and verifying its own training data. Based on the established high-throughput experimental simulation framework, the generated simulated experimental data set, and the deep neural network model established based on the Python environment and Tensorlow library, a dynamic iterative active machine learning framework is constructed.

[0101] Piper three-line diagram Figure 4 As shown, the piper three-line diagram is composed based on the normalization of key components (the key components of biomass are cellulose, hemicellulose, and lignocellulose, and the key components of plastics are the contents of PP, LDPE, and PS), and the data point distribution is obtained using the trained deep neural network model.

[0102] The piper trigraph consists of two separate ternary graphs, which are drawn based on the normalized composition of the key components. One reflects the content of cellulose, hemicellulose and lignocellulose in the biomass, while the other represents the content of PP, LDPE and PS.

[0103] The relationship between biomass content and plastic content is projected together into a diamond diagram. The diamond diagram consists of two main axes, parallel to the ternary diagram and with the same scale as the axes of the ternary diagram. The samples plotted on the ternary diagram of biomass chemical composition and plastic composition project a point on the diamond diagram for analyzing the co-pyrolysis products and environmental impacts of biomass and plastics.

[0104] First, the deep neural network model trained on the original simulated experimental data set will accurately classify the new experimental research data samples regarding product classification, and generate additional biomass co-pyrolysis data through simulation and experimental verification of new combinations of biomass and plastic reactants.

[0105] Adding additional data points to the original simulated experiment dataset and then retraining the deep neural network model with the new dataset containing the validation data to update the deep neural network's prediction and classification priorities will produce new and more accurate predictions for the same data points.

[0106] Therefore, through coupling with experiments, the deep neural network model continuously updates and verifies its own training data, thus forming a high-throughput biomass pyrolysis active learning framework that is automatically and systematically updated and iterated.

[0107] By continuously and automatically executing deep neural network classification to guide the accurate classification of biomass co-pyrolysis product distribution, the original SPD kinetic model and WAR evaluation are optimized, and the deep machine learning classification boundaries of the generated piper three-line diagram are dynamically adjusted and optimized to further clarify the correspondence between reactant-product results and achieve bidirectional and accurate prediction of reactant-product combinations.

[0108] Accordingly, the present application provides an electronic device, comprising: a processor; a memory for storing instructions executable by the processor; wherein the processor is configured to execute the above-mentioned agricultural and plastic waste co-pyrolysis process prediction and optimization platform. Figure 5 As shown, the processor and the memory are connected via a bus, and the memory stores instructions that can be executed by the processor. The instructions are executed by the processor so that the electronic device can execute the entire process of the agricultural and plastic waste co-pyrolysis process prediction method. Figure 5 These are merely examples of electronic devices and do not constitute a limitation of the electronic device. The electronic device may include more or fewer components than those shown in the figure, or a combination of certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0109] Figure 6The present invention is a schematic diagram of the structure of a device for predicting the co-pyrolysis products of biomass and plastics, including a display screen and an information input component. The display screen is used to display the data output by the agricultural and plastic waste co-pyrolysis process prediction and optimization platform, and the information input component is used for users to input data into the agricultural and plastic waste co-pyrolysis process prediction and optimization platform. The information input component can be a keyboard, and the display screen and the information input component can also be a touch screen or a mobile computer with an integrated structure. The display screen can display information including but not limited to: prediction and environmental assessment of the input biomass-plastic co-pyrolysis products, and the types of biomass and plastic ratios obtained by inversely deducing the expected products and environmental requirements.

[0110] The purpose of this application is to predict the products and environmental impacts of co-pyrolysis of known biomass and plastics through a combination of experiments and simulations, and to infer the required biomass type and plastic composition from the required products.

[0111] The specific advantages are:

[0112] This application combines the biomass cell structure and the components of cellulose, hemicellulose, and lignocellulose to study the biomass pyrolysis reaction path and reveal its reaction mechanism; it studies the effects of biomass components and the ratio of three plastics on the co-pyrolysis products and reveals the co-pyrolysis mechanism of biomass and plastics.

[0113] This application combines experiments and high-throughput simulations to carry out efficient biomass-plastic co-pyrolysis product prediction and environmental assessment, builds a one-stop platform in the form of a piper three-line diagram, and uses active learning algorithms to continuously optimize and modify the platform.

[0114] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the above-mentioned method for predicting co-pyrolysis products of biomass and plastic when executing the computer program.

[0115] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, which, when executed by a processor, implements the above-mentioned method for predicting co-pyrolysis products of biomass and plastic.

[0116] In an exemplary embodiment, a computer program product is provided, including a computer program, which, when executed by a processor, implements the above-mentioned method for predicting co-pyrolysis products of biomass and plastic.

[0117] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 7As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for predicting co-pyrolysis products of biomass and plastics is implemented.

[0118] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0119] 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, stored data, displayed data, 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 relevant data must comply with relevant regulations.

[0120] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present 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 may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0121] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0122] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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.

[0123] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for predicting products of co-pyrolysis of biomass and plastics, characterized in that: include: Obtain the types of biomass to be decomposed and the proportion of plastic composition; According to the types of biomass to be decomposed and the composition ratio of plastics, the co-pyrolysis product prediction model is used to determine the co-pyrolysis product ratio; wherein the co-pyrolysis product prediction model is obtained by training a deep neural network model using a training data set; the training data set is based on the types of training biomass and the composition ratio of training plastics, and is determined by high-throughput experimental simulation; the training data set includes the types of training biomass, the composition ratio of training plastics and the corresponding co-pyrolysis product ratios.

2. The method for predicting co-pyrolysis products of biomass and plastics according to claim 1, characterized in that: According to the type of biomass to be decomposed and the proportion of plastic composition, the co-pyrolysis product prediction model is used to determine the proportion of co-pyrolysis products, including: Standardizing the types of biomass to be decomposed and the composition ratio of plastics to obtain the types of biomass after treatment and the composition ratio of plastics after treatment; According to the type of biomass after treatment and the composition ratio of plastic after treatment, the ratio of co-pyrolysis products is determined by using the co-pyrolysis product prediction model.

3. The method for predicting the co-pyrolysis products of biomass and plastic according to claim 1, characterized in that: Based on the types of biomass used for training and the proportion of plastic used for training, the training data set is determined by high-throughput experimental simulation, including: Construct the co-pyrolysis reaction mechanism of biomass and plastics; the co-pyrolysis reaction mechanism of biomass and plastics is the influence of biomass type and plastic composition ratio on co-pyrolysis products; According to the co-pyrolysis reaction mechanism of biomass and plastics, a kinetic model of the co-pyrolysis reaction of biomass and plastics was constructed; Based on the training biomass types, training plastic composition ratios, and the kinetic model of the co-pyrolysis reaction of biomass and plastics, the SuperPro Designer software was used to simulate the reaction process, and the component object model was used to drive the high-throughput experimental simulation to determine the co-pyrolysis products and the co-pyrolysis product ratios, thereby constructing a high-throughput biomass pyrolysis data set; Preprocessing the high-throughput biomass pyrolysis data set to obtain a processed high-throughput biomass pyrolysis data set; The processed high-throughput biomass pyrolysis dataset is divided into training dataset, validation dataset and test dataset.

4. The method for predicting the co-pyrolysis products of biomass and plastic according to claim 3, characterized in that: Also includes: Based on a high-throughput biomass pyrolysis dataset, a waste reduction algorithm was used to determine the environmental impact of co-pyrolysis of biomass and plastics. The environmental impacts of co-pyrolysis of biomass and plastics include CO2 emissions.

5. The method for predicting the co-pyrolysis products of biomass and plastics according to claim 4, characterized in that: Also includes: According to the type of biomass to be decomposed and the proportion of plastic composition, the co-pyrolysis product prediction model is used to determine the impact of co-pyrolysis of biomass and plastic on the environment.

6. The method for predicting the co-pyrolysis products of biomass and plastics according to claim 1, characterized in that: Use the training data set to train the deep neural network model, including: Taking the types of processed training biomass and the composition ratio of processed training plastics in the current training data set as input and the corresponding co-pyrolysis product ratios in the current training data set as output, the current deep neural network model is trained to obtain a trained deep neural network model; The trained deep neural network model is verified and tested using the verification data set and the test data set respectively to obtain the initial deep neural network model; Obtain the types of biomass samples and the composition ratio of plastic samples, use the current initial deep neural network model to determine the ratio of co-pyrolysis products of biomass samples and plastic samples, and draw a piper three-line diagram; Determine whether the current initial deep neural network model reaches the preset prediction accuracy; If yes, the current initial deep neural network model is used as the co-pyrolysis product prediction model; If not, based on the types of biomass samples, the composition ratio of plastic samples, and the kinetic model of the co-pyrolysis reaction of biomass and plastics, the SuperPro Designer software is used to simulate the reaction process, and the component object model is used to drive the high-throughput experimental simulation to determine the co-pyrolysis products and the ratio of the co-pyrolysis products, and the current training data set is expanded to obtain the expanded training data set; Take the expanded training data set as the current data set, and return "Use the types of processed training biomass and the composition ratio of processed training plastics in the current training data set as input, and the corresponding co-pyrolysis product ratios in the current training data set as output, train the current deep neural network model, and obtain the trained deep neural network model".

7. A computer device comprising: 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 method for predicting co-pyrolysis products of biomass and plastics according to any one of claims 1 to 6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for predicting the co-pyrolysis products of biomass and plastics according to any one of claims 1 to 6 is implemented.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for predicting the co-pyrolysis products of biomass and plastics according to any one of claims 1 to 6 is implemented.

Images