Waste plastic catalytic conversion cascade system and method based on artificial intelligence

Abstract

The invention provides a waste plastic catalytic conversion cascade system and method based on artificial intelligence. The waste plastic catalytic conversion cascade system and method based on artificial intelligence comprises an artificial intelligence sorting system used for collecting physical and chemical characteristic data of waste plastics in real time and accurately classifying the waste plastics through image recognition and a deep learning algorithm; and the hydrolysis reactor is used for carrying out an alkali thermal hydrolysis reaction under an alkaline condition and converting the waste plastic into an intermediate product which can be further processed. According to the waste plastic catalytic conversion cascade system and method based on artificial intelligence, an artificial intelligence sorting system is introduced, physical and chemical characteristics of waste plastic are collected in real time through image recognition and a deep learning algorithm, and accurate classification is conducted. According to the system, corresponding treatment channels and reaction conditions can be automatically selected and adjusted according to the types of waste plastics, and it is ensured that different types of waste plastics are treated most appropriately.

Description

Technical Field

[0001] The present invention relates to the technical field of waste plastic processing and conversion, and specifically to an artificial intelligence-based waste plastic catalytic conversion cascade system and method. Background Art

[0002] The disposal of waste plastics has become a major global environmental challenge, particularly given the widespread use of plastics. Traditional methods for processing waste plastics, such as mechanical recycling and pyrolysis, often suffer from low energy efficiency, poor conversion rates, and pollutant emissions, resulting in poor resource recovery and potential environmental harm. Therefore, improving the conversion efficiency of waste plastics, reducing energy consumption, and achieving sustainable utilization have become key research priorities.

[0003] In recent years, thermal catalysis and electrocatalysis technologies have gradually become important means of treating waste plastics, especially in the field of catalytic reactions, showing significant application potential. By combining thermal catalysis with electrocatalysis, the conversion rate of waste plastics can be significantly improved to generate high-value-added chemicals and energy. However, there is still much room for improvement in the existing technology in terms of catalyst selection, optimization of reaction processes, and accurate classification of waste plastics. The diversity of waste plastics means that different types of plastics have different requirements for reaction conditions and catalysts during the conversion process. Existing technologies cannot meet the needs of efficient conversion of different types of waste plastics.

[0004] Therefore, the development of a waste plastic catalytic conversion cascade system that integrates intelligent classification, catalyst screening, and reaction condition adjustment has important theoretical significance and practical application value for improving the efficiency of waste plastic resource conversion and product quality. Summary of the Invention

[0005] In response to the shortcomings of the existing technology, the present invention provides an artificial intelligence-based catalytic conversion cascade system and method for waste plastics, which solves the problems of catalyst selection, reaction process and insufficient precision of waste plastics.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: an artificial intelligence-based waste plastic catalytic conversion cascade system, comprising:

[0007] An artificial intelligence sorting system collects physical and chemical characteristic data of waste plastics in real time, accurately classifies waste plastics through image recognition and deep learning algorithms, and distributes waste plastics to different processing channels according to their type;

[0008] A hydrolysis reactor for performing an alkaline-thermal hydrolysis reaction under alkaline conditions to convert waste plastics into intermediate products that can be further processed. The reaction conditions of the alkaline-thermal hydrolysis reaction, including temperature and pH, are automatically controlled according to the type of waste plastics;

[0009] Electrocatalytic reactors, which are used to convert the intermediate products generated by the hydrolysis reaction into high-value-added chemicals and green hydrogen through electrocatalytic reactions, where each electrocatalytic reactor is dedicated to processing one type of waste plastic;

[0010] The electrocatalyst screening module uses an artificial intelligence sorting system to automatically select suitable electrocatalysts based on the type and characteristics of waste plastics. Each processing channel corresponds to a specific electrocatalyst, ensuring optimized reaction efficiency and improved selectivity of the catalytic reaction. The selected catalyst is automatically matched with the corresponding electrocatalytic reactor based on the type and characteristics of the waste plastics.

[0011] Preferably, the artificial intelligence sorting system includes several sensor arrays, high-definition cameras, infrared sensors and data processing modules. The artificial intelligence sorting system captures the morphology, surface texture, size, color and physical properties of waste plastics in real time. The data processing module uses a convolutional neural network to process the data, accurately identify the type of waste plastics, and classify the waste plastics into different processing channels, so that subsequent processing modules can select appropriate hydrolysis and electrocatalytic reaction conditions according to the type of waste plastics.

[0012] Preferably, the hydrolysis reactor adopts a circulating flow reactor design and is combined with a heat recovery system to collect waste heat generated during the hydrolysis process through an efficient heat exchanger, transfer the heat to the reaction zone, maintain a stable reaction temperature, thereby improving reaction efficiency and reducing the demand for external energy.

[0013] Preferably, the hydrolysis reactor includes a thermal gradient reactor, which optimizes the reaction temperature distribution by setting different temperature control zones, thereby assisting the artificial intelligence sorting system in accurately controlling the reaction temperature, further improving the efficiency of the catalytic reaction, and enhancing the waste plastic conversion rate and product selectivity.

[0014] Preferably, the electrocatalytic reactor adjusts the applied potential in real time to adapt to the conversion requirements of different types of waste plastics, thereby maximizing the efficiency of the catalytic reaction. The artificial intelligence sorting system adjusts the operating parameters of the electrocatalytic reactor according to real-time reaction data to improve the generation efficiency of green hydrogen and ensure the efficient generation of high value-added chemicals.

[0015] Preferably, the type and performance of the electrocatalyst in the electrocatalytic reactor are pre-set according to the type of waste plastic. Each electrocatalytic reactor corresponds to a specific type of waste plastic, and according to the sorting results of the waste plastic, the artificial intelligence sorting system allocates the waste plastic to different electrocatalytic treatment channels. The electrocatalyst uses foamed metal nickel or foamed carbon as a substrate, and the load includes but is not limited to iron, cobalt, and nickel. The selection of the electrocatalyst is determined by the artificial intelligence sorting system according to the sorting type of the waste plastic, ensuring efficient electrolytic conversion of the waste plastic in a specific reactor to generate specific target chemicals.

[0016] Preferably, the artificial intelligence sorting system can not only classify waste plastics before they enter the reaction system, but also monitor the reaction conditions in real time and dynamically adjust the control strategy according to the reaction progress. The artificial intelligence sorting system uses deep learning algorithms or optimization control algorithms, combined with real-time reaction data, to predict and adjust the operating conditions of each reaction stage to ensure that each step in the reaction process is in the optimal state, thereby improving conversion efficiency and product selectivity.

[0017] Preferably, the reaction rate equation of the hydrolysis reaction is:

[0018] r hydrolysis =k hydrolysis ·[plastic]·[OH - ]

[0019] in:

[0020] r hydrolysis is the rate of the hydrolysis reaction (mol / L·h)

[0021] The k hydrolysis is the rate constant of the hydrolysis reaction, which is affected by temperature and pH and can be expressed as the Arrhenius equation:

[0022]

[0023] A is the frequency factor;

[0024] E a is the activation energy (J / mol);

[0025] R is the gas constant (8.314 J / mol·K);

[0026] T is the reaction temperature (K);

[0027] [plastic] is the concentration of waste plastic (mol / L);

[0028] [OH -] is the concentration of hydroxide ions (mol / L), which is determined by the alkaline environment of the hydrolysis reaction and the pH value is controlled between 10 and 12;

[0029] Effects of temperature and pH on the reaction:

[0030] T=T set

[0031] pH=10-12

[0032] By adjusting the temperature and pH, the rate and efficiency of the hydrolysis reaction can be optimized.

[0033] Preferably, the convolutional neural network model is used to process image data of waste plastics and accurately identify the type of plastics. The design of the convolutional neural network model includes the training of the convolutional neural network structure, loss function and optimizer and the convolutional neural network model.

[0034] The convolutional neural network structure

[0035] Input layer: image data, appearance images of discarded plastics, input is a fixed size of 256x256 pixels.

[0036] Convolutional layer:

[0037] The first convolutional layer uses multiple 3x3 convolution kernels to extract low-level features of the image, including edges and corners;

[0038] The second convolutional layer: extracts more advanced features, such as shape and texture;

[0039] Pooling layer: downsample the convolution features to reduce the amount of calculation, using a maximum pooling layer with a size of 2x2;

[0040] Fully connected layer: flattens the features extracted by the convolutional layer and passes them to the fully connected layer for classification;

[0041] Output layer: outputs a multi-class classification result based on the number of categories of waste plastics;

[0042] The loss function and optimizer use the cross entropy loss function, which is suitable for multi-category classification tasks:

[0043]

[0044] in:

[0045] y i is the actual label, the true value,

[0046] p i is the predicted probability;

[0047] Optimizer: Use Adam optimizer to update network parameters:

[0048]

[0049] in:

[0050] θ t are the current network parameters,

[0051] m t and v t are the first and second moment estimates,

[0052] η is the learning rate, ∈ is a small constant to prevent division by zero errors;

[0053] During the training process of the convolutional neural network model, the classification accuracy is optimized by continuously adjusting the weights. Each time an image of discarded plastic is input, the network weights are gradually optimized according to the label of the actual category to ensure classification accuracy.

[0054] An artificial intelligence-based intelligent sorting and catalytic conversion method for waste plastics, comprising the following steps:

[0055] Step 1: Waste plastics are intelligently classified through an artificial intelligence system, which collects real-time data on their physical and chemical properties, including but not limited to morphology, surface texture, size, color, and chemical composition. The system then uses a deep learning algorithm to analyze the type of waste plastics and classify them into corresponding processing channels, allowing subsequent processing modules to select appropriate conversion conditions based on the type of waste plastics.

[0056] Step 2: The sorted waste plastics are fed into a hydrolysis reactor and reacted with a catalyst under alkaline conditions and thermal catalytic conditions to convert them into intermediates. The temperature, pH value, catalyst type and reaction time of the hydrolysis reaction process are adjusted according to the type of waste plastics and reactor conditions to maximize the conversion efficiency and optimize the formation of intermediates. The temperature range of the hydrolysis reaction is 140°C to 280°C, and the pH value is controlled between 10 and 12 to ensure efficient decomposition of the waste plastics.

[0057] Step 3: The intermediates are transported via a pipeline to an electrocatalytic reactor for further conversion under electrocatalytic conditions. Based on the type of waste plastic and the characteristics of the intermediates generated by the hydrolysis reaction, the artificial intelligence sorting system automatically selects a suitable electrocatalyst and adjusts the current and voltage based on the type of plastic, the reaction progress, and real-time reaction data in the electrolysis cell. The electrocatalyst includes a nickel foam or a carbon foam as a substrate, loaded with iron, cobalt, and nickel metal elements to optimize reaction efficiency and ensure reaction selectivity, further converting the intermediates into high-value-added chemicals and green hydrogen.

[0058] Step 4: The AI control system monitors and adjusts the temperature, pressure, and applied potential in the reactor in real time. The AI sorting system dynamically adjusts the reaction parameters based on real-time feedback data to ensure that each step in the reaction process is always in the optimal state, maximizing the conversion efficiency of waste plastics and the quality of high-value-added products. The AI sorting system continuously monitors the reaction data to optimize the temperature and current density of each reaction stage, thereby improving the reaction stability and product selectivity.

[0059] Step 5: Separate and collect the generated high-value-added chemicals and green hydrogen, and use them for industrial applications or energy production respectively. For example, the generated green hydrogen can be used as a clean energy to replace traditional fossil energy; the high-value-added chemicals can be used in chemical raw materials, material recycling and other fields to further improve resource utilization.

[0060] The present invention provides an artificial intelligence-based catalytic conversion cascade system and method for waste plastics. It has the following beneficial effects:

[0061] This AI-based catalytic conversion cascade system and method for waste plastics utilizes an AI sorting system, image recognition, and deep learning algorithms to collect the physical and chemical characteristics of waste plastics in real time and accurately classify them. This system automatically selects and adjusts the appropriate processing channels and reaction conditions based on the type of waste plastic, ensuring that different types of waste plastic receive the most appropriate treatment. This not only improves processing efficiency but also ensures the efficient conversion of waste plastics, resolving the issues of inaccurate waste plastic classification and mismatched reaction conditions encountered in traditional technologies.

[0062] The design of the hydrolysis reactor and electrocatalytic reactor fully considers the type of waste plastics and the reaction requirements. By combining a heat recovery system and an electrocatalyst screening module, the system can optimize reaction conditions such as temperature, pH value, and current at different reaction stages, ensuring the efficient conversion of waste plastics and maximizing the production of green hydrogen and high-value-added chemicals. In addition, the artificial intelligence sorting system not only classifies waste plastics before the reaction, but also monitors and adjusts reaction parameters in real time, improving the stability of the reaction process and the selectivity of the products. Through this integrated solution, the efficiency of the catalytic reaction and the quality of the products are significantly improved, the demand for external energy is reduced, and resource utilization is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 Schematic diagram of the internal structure of the present invention;

[0064] Figure 2 Schematic diagram of the present invention. DETAILED DESCRIPTION

[0065] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0066] like Figure 1-2 As shown, an embodiment of the present invention provides a catalytic conversion cascade system for waste plastics based on artificial intelligence, including an artificial intelligence sorting system for real-time collection of physical and chemical characteristic data of waste plastics, and accurate classification of waste plastics through image recognition and deep learning algorithms, and allocation of waste plastics according to their types to different processing channels. The artificial intelligence sorting system includes several sensor arrays, high-definition cameras, infrared sensors and data processing modules. The artificial intelligence sorting system captures the shape, surface texture, size, color and physical properties of waste plastics in real time. The data processing module uses a convolutional neural network to process the data, accurately identify the type of waste plastics, and classify the waste plastics into different processing channels for subsequent processing. The processing module selects appropriate hydrolysis and electrocatalytic reaction conditions according to the type of waste plastics. The artificial intelligence sorting system can not only classify the waste plastics before they enter the reaction system, but also monitor the reaction conditions in real time and dynamically adjust the control strategy according to the reaction progress. The artificial intelligence sorting system uses deep learning algorithms or optimization control algorithms, combined with real-time reaction data, to predict and adjust the operating conditions of each reaction stage to ensure that each step in the reaction process is in the optimal state, improve conversion efficiency and product selectivity, and the convolutional neural network model is used to process image data of waste plastics and accurately identify the type of plastic. The design of the convolutional neural network model includes the convolutional neural network structure, loss function and optimizer, and the training of the convolutional neural network model.

[0067] Convolutional Neural Network Architecture

[0068] Input layer: image data, appearance images of discarded plastics, input is a fixed size of 256x256 pixels.

[0069] Convolutional layer:

[0070] The first convolution layer uses multiple 3x3 convolution kernels to extract low-level features of the image, including edges and corners.

[0071] The second convolutional layer: extracts more advanced features, such as shape and texture.

[0072] Pooling layer: downsample the convolution features to reduce the amount of calculation, using a maximum pooling layer with a size of 2x2.

[0073] Fully connected layer: The features extracted by the convolutional layer are flattened and passed to the fully connected layer for classification.

[0074] Output layer: Output a multi-class classification result based on the number of categories of waste plastics.

[0075] The loss function and optimizer use the cross entropy loss function, which is suitable for multi-category classification tasks:

[0076]

[0077] in:

[0078] y i is the actual label, the true value,

[0079] p i is the predicted probability.

[0080] Optimizer: Use Adam optimizer to update network parameters:

[0081]

[0082] in:

[0083] θ t are the current network parameters,

[0084] m t and v t are the first and second moment estimates,

[0085] η is the learning rate and ∈ is a small constant to prevent division by zero errors.

[0086] During the training process of the convolutional neural network model, the weights are continuously adjusted to optimize the classification accuracy. Each time an image of discarded plastic is input, the network weights will be gradually optimized according to the label of the actual category to ensure classification accuracy.

[0087] The hydrolysis reactor is used to carry out alkaline-thermal hydrolysis under alkaline conditions to convert waste plastics into intermediate products that can be further processed. The reaction conditions of the alkaline-thermal hydrolysis reaction, including temperature and pH value, are automatically adjusted according to the type of waste plastics. The hydrolysis reactor adopts a circulating flow reactor design and is combined with a heat recovery system. The waste heat generated during the hydrolysis process is collected through an efficient heat exchanger and transferred to the reaction zone to maintain a stable reaction temperature, thereby improving reaction efficiency and reducing the demand for external energy. The hydrolysis reactor includes a thermal gradient reactor. The thermal gradient reactor optimizes the reaction temperature distribution by setting different temperature control zones, thereby assisting the artificial intelligence sorting system in accurately controlling the reaction temperature, further improving the efficiency of the catalytic reaction, and enhancing the waste plastic conversion rate and product selectivity. The reaction rate equation of the hydrolysis reaction is:

[0088] rhydrolysis =k hydrolysis ·[plastic]·[OH - ]

[0089] in:

[0090] r hydrolysis is the rate of the hydrolysis reaction (mol / L·h)

[0091] k hydrolysis is the rate constant of the hydrolysis reaction, which is affected by temperature and pH and can be expressed as the Arrhenius equation:

[0092]

[0093] A is the frequency factor.

[0094] E a is the activation energy (J / mol).

[0095] R is the gas constant (8.314 J / mol·K).

[0096] T is the reaction temperature (K).

[0097] [plastic] is the concentration of waste plastic (mol / L).

[0098] [OH - ] is the concentration of hydroxide ions (mol / L), which is determined by the alkaline environment of the hydrolysis reaction and the pH value is controlled between 10 and 12.

[0099] Effects of temperature and pH on the reaction:

[0100] T=T set

[0101] pH=10-12

[0102] By adjusting the temperature and pH, the rate and efficiency of the hydrolysis reaction can be optimized.

[0103] The electrocatalytic reactor is used to convert the intermediate products generated by the hydrolysis reaction into high-value-added chemicals and green hydrogen through electrocatalytic reactions. Each electrocatalytic reactor specializes in processing one type of waste plastic. The electrocatalytic reactor adjusts the external potential in real time to adapt to the conversion requirements of different types of waste plastics, thereby maximizing the efficiency of the catalytic reaction. The artificial intelligence sorting system adjusts the operating parameters of the electrocatalytic reactor according to the real-time reaction data to improve the generation efficiency of green hydrogen and ensure the efficient generation of high-value-added chemicals. The type and performance of the electrocatalyst in the electrocatalytic reactor are pre-set according to the type of waste plastic. Each electrocatalytic reactor corresponds to a specific type of waste plastic, and according to the sorting results of the waste plastic, the artificial intelligence sorting system allocates the waste plastic to different electrocatalytic treatment channels. The electrocatalyst uses foamed metal nickel or foamed carbon as the base, and the load includes but is not limited to iron, cobalt, and nickel. The choice of electrocatalyst is determined by the artificial intelligence sorting system according to the sorting type of the waste plastic to ensure efficient electrolytic conversion of the waste plastic in the specific reactor to generate specific target chemicals.

[0104] The electrocatalyst screening module uses an artificial intelligence sorting system to automatically select suitable electrocatalysts based on the type and characteristics of waste plastics. Each processing channel corresponds to a specific electrocatalyst, ensuring optimized reaction efficiency and improved selectivity of the catalytic reaction. The selected catalyst is automatically matched with the corresponding electrocatalytic reactor based on the type and characteristics of the waste plastics.

[0105] An artificial intelligence-based intelligent sorting and catalytic conversion method for waste plastics, comprising the following steps:

[0106] Step 1: Waste plastics are intelligently classified through an artificial intelligence system, which collects real-time data on their physical and chemical properties, including but not limited to morphology, surface texture, size, color, and chemical composition. The system then uses a deep learning algorithm to analyze the type of waste plastics and classify them into corresponding processing channels, allowing subsequent processing modules to select appropriate conversion conditions based on the type of waste plastics.

[0107] Step 2: The sorted waste plastics are fed into a hydrolysis reactor and reacted with a catalyst under alkaline conditions and thermal catalytic conditions to convert them into intermediates. The temperature, pH value, catalyst type and reaction time of the hydrolysis reaction process are adjusted according to the type of waste plastics and reactor conditions to maximize the conversion efficiency and optimize the formation of intermediates. The temperature range of the hydrolysis reaction is 140°C to 280°C, and the pH value is controlled between 10 and 12 to ensure efficient decomposition of the waste plastics.

[0108] Step 3: The intermediates are transported via a pipeline to an electrocatalytic reactor for further conversion under electrocatalytic conditions. Based on the type of waste plastic and the characteristics of the intermediates generated by the hydrolysis reaction, the artificial intelligence sorting system automatically selects a suitable electrocatalyst and adjusts the current and voltage based on the type of plastic, the reaction progress, and real-time reaction data in the electrolysis cell. The electrocatalyst includes a nickel foam or a carbon foam as a substrate, loaded with iron, cobalt, and nickel metal elements to optimize reaction efficiency and ensure reaction selectivity, further converting the intermediates into high-value-added chemicals and green hydrogen.

[0109] Step 4: The AI control system monitors and adjusts the temperature, pressure, and applied potential in the reactor in real time. The AI sorting system dynamically adjusts the reaction parameters based on real-time feedback data to ensure that each step in the reaction process is always in the optimal state, maximizing the conversion efficiency of waste plastics and the quality of high-value-added products. The AI sorting system continuously monitors the reaction data to optimize the temperature and current density of each reaction stage, thereby improving the reaction stability and product selectivity.

[0110] Step 5: Separate and collect the generated high-value-added chemicals and green hydrogen, and use them for industrial applications or energy production respectively. For example, the generated green hydrogen can be used as a clean energy to replace traditional fossil energy; the high-value-added chemicals can be used in chemical raw materials, material recycling and other fields to further improve resource utilization.

[0111] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A cascade system and method for catalytic conversion of waste plastics based on artificial intelligence, characterized in that: include: An artificial intelligence sorting system is used to collect physical and chemical characteristic data of waste plastics in real time and accurately classify waste plastics through image recognition and deep learning algorithms; A hydrolysis reactor for performing an alkaline-thermal hydrolysis reaction under alkaline conditions to convert waste plastics into intermediate products that can be further processed; Electrocatalytic reactors, which are used to convert the intermediate products generated by the hydrolysis reaction into high-value-added chemicals and green hydrogen through electrocatalytic reactions, where each electrocatalytic reactor is dedicated to processing one type of waste plastic; The electrocatalyst screening module uses an artificial intelligence sorting system to automatically select suitable electrocatalysts based on the type and characteristics of waste plastics. Each processing channel corresponds to a specific electrocatalyst, ensuring optimized reaction efficiency and improved selectivity of the catalytic reaction. The selected catalyst is automatically matched with the corresponding electrocatalytic reactor based on the type and characteristics of the waste plastics.

2. The artificial intelligence-based waste plastic catalytic conversion cascade system according to claim 1, characterized in that: The artificial intelligence sorting system includes several sensor arrays, high-definition cameras, infrared sensors and data processing modules. The artificial intelligence sorting system captures the shape, surface texture, size, color and physical properties of waste plastics in real time. The data processing module uses a convolutional neural network to process the data.

3. The artificial intelligence-based waste plastic catalytic conversion cascade system according to claim 1, characterized in that: The hydrolysis reactor adopts a circulating flow reactor design and is combined with a heat recovery system.

4. The artificial intelligence-based waste plastic catalytic conversion cascade system according to claim 1, characterized in that: The hydrolysis reactor includes a thermal gradient reactor, which optimizes the reaction temperature distribution by setting different temperature control areas, thereby assisting the artificial intelligence sorting system in accurately controlling the reaction temperature.

5. The artificial intelligence-based waste plastic catalytic conversion cascade system according to claim 1, characterized in that: The electrocatalytic reactor adjusts the applied potential in real time, and the artificial intelligence sorting system adjusts the operating parameters of the electrocatalytic reactor according to real-time reaction data.

6. The artificial intelligence-based waste plastic catalytic conversion cascade system according to claim 1, characterized in that: The type and performance of the electrocatalyst in the electrocatalytic reactor are pre-set according to the type of waste plastic. The electrocatalyst uses foamed metal nickel or foamed carbon as a substrate, and the load includes but is not limited to iron, cobalt, and nickel. The selection of the electrocatalyst is determined by the artificial intelligence sorting system according to the sorting type of the waste plastic.

7. The artificial intelligence-based waste plastic catalytic conversion cascade system according to claim 1, characterized in that: The artificial intelligence sorting system uses deep learning algorithms or optimization control algorithms, combined with real-time reaction data, to predict and adjust the operating conditions of each reaction stage to ensure that each step in the reaction process is in the optimal state.

8. The artificial intelligence-based waste plastic catalytic conversion cascade system according to claim 1, characterized in that: The reaction rate equation of the hydrolysis reaction is: r hydrolysis =k hydrolysis ·[plastic]·[OH - ] in: r hydrolysis is the rate of the hydrolysis reaction (mol / L·h) The k hydrolysis is the rate constant of the hydrolysis reaction, which is affected by temperature and pH and can be expressed as the Arrhenius equation: A is the frequency factor; E a is the activation energy (J / mol); R is the gas constant (8.314 J / mol·K); T is the reaction temperature (K); [plastic] is the concentration of waste plastic (mol / L); [OH - ] is the concentration of hydroxide ions (mol / L), which is determined by the alkaline environment of the hydrolysis reaction and the pH value is controlled between 10 and 12; Effects of temperature and pH on the reaction: T=T set pH=10-12 By adjusting the temperature and pH, the rate and efficiency of the hydrolysis reaction can be optimized.

9. The artificial intelligence-based waste plastic catalytic conversion cascade system according to claim 1, characterized in that: The convolutional neural network model is used to process image data of waste plastics and accurately identify the type of plastics. The design of the convolutional neural network model includes the convolutional neural network structure, loss function and optimizer, and training of the convolutional neural network model. The convolutional neural network structure Input layer: image data, appearance images of discarded plastics, input is a fixed size of 256x256 pixels. Convolutional layer: The first convolutional layer uses multiple 3x3 convolution kernels to extract low-level features of the image, including edges and corners; The second convolutional layer: extracts more advanced features, such as shape and texture; Pooling layer: downsample the convolution features to reduce the amount of calculation, using a maximum pooling layer with a size of 2x2; Fully connected layer: flattens the features extracted by the convolutional layer and passes them to the fully connected layer for classification; Output layer: outputs a multi-class classification result based on the number of categories of waste plastics; The loss function and optimizer use the cross entropy loss function, which is suitable for multi-category classification tasks: in: y i is the actual label, the true value, p i is the predicted probability; Optimizer: Use Adam optimizer to update network parameters: in: θ t are the current network parameters, m t and v t are the first and second moment estimates, η is the learning rate, ∈ is a small constant to prevent division by zero errors; During the training process of the convolutional neural network model, the classification accuracy is optimized by continuously adjusting the weights. Each time an image of discarded plastic is input, the network weights are gradually optimized according to the label of the actual category to ensure classification accuracy.

10. An artificial intelligence-based intelligent sorting and catalytic conversion method for waste plastics, characterized in that: The following steps are involved: Step 1: Waste plastics are intelligently classified through an artificial intelligence system, which collects real-time data on their physical and chemical properties, including but not limited to morphology, surface texture, size, color, and chemical composition. The system then uses a deep learning algorithm to analyze the type of waste plastics and classify them into corresponding processing channels, allowing subsequent processing modules to select appropriate conversion conditions based on the type of waste plastics. Step 2: The sorted waste plastics are fed into a hydrolysis reactor and reacted with a catalyst under alkaline conditions and thermal catalytic conditions to convert them into intermediates. The temperature, pH value, catalyst type and reaction time of the hydrolysis reaction process are adjusted according to the type of waste plastics and reactor conditions to maximize the conversion efficiency and optimize the formation of intermediates. The temperature range of the hydrolysis reaction is 140°C to 280°C, and the pH value is controlled between 10 and 12 to ensure efficient decomposition of the waste plastics. Step 3: The intermediates are transported via a pipeline to an electrocatalytic reactor for further conversion under electrocatalytic conditions. Based on the type of waste plastic and the characteristics of the intermediates generated by the hydrolysis reaction, the artificial intelligence sorting system automatically selects a suitable electrocatalyst and adjusts the current and voltage based on the type of plastic, the reaction progress, and real-time reaction data in the electrolysis cell. The electrocatalyst includes a nickel foam or a carbon foam as a substrate, loaded with iron, cobalt, and nickel metal elements to optimize reaction efficiency and ensure reaction selectivity, further converting the intermediates into high-value-added chemicals and green hydrogen. Step 4: The AI control system monitors and adjusts the temperature, pressure, and applied potential in the reactor in real time. The AI sorting system dynamically adjusts the reaction parameters based on real-time feedback data to ensure that each step in the reaction process is always in the optimal state, maximizing the conversion efficiency of waste plastics and the quality of high-value-added products. Step 5: Separate and collect the generated high-value-added chemicals and green hydrogen for use in industrial applications or energy production respectively.

Images