Carbon capture system based on artificial intelligence optimization control

Abstract

The invention belongs to the technical field of carbon capture, and provides a carbon capture system based on artificial intelligence optimization control, the carbon capture system comprises a flue gas pipeline, carbon capture films, a sensing module, a control module and an execution module, the flue gas pipeline is internally provided with a plurality of carbon capture films, the pore size, pore size distribution, porosity, permeability and thickness of the carbon capture films can be adjusted, and the sensing module is connected with the control module. A sensing module is arranged in front of each carbon capture film, the sensing modules are used for collecting operation data in the flue gas pipeline, the sensing modules are connected with the input end of the control module and transmit the operation data in the flue gas pipeline to the control module, and the output end of the control module is connected with the execution module. The control module is used for analyzing and processing according to the operation data to obtain an adjusting instruction and sending the adjusting instruction to the execution module, the execution module is arranged in the flue gas pipeline, and the execution module adjusts the temperature, humidity and pressure of gas in the flue gas pipeline according to the adjusting instruction. According to the invention, the CO2 capture efficiency and selectivity can be improved, and the operation energy consumption can be reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of carbon capture, and specifically discloses a carbon capture system based on artificial intelligence optimized control.

[0002] Background Introduction

[0003] With the increasingly serious problem of global warming, carbon capture technology has become an important way to reduce CO 2 emissions. However, traditional carbon capture technologies face problems such as low selectivity, high energy consumption, and complex maintenance, and cannot meet the requirements of large-scale industrial applications. Therefore, developing an efficient, intelligent, and low-energy carbon capture system has become an important direction for technological development. Currently, the problem of global climate change is becoming increasingly serious, and the excessive emission of greenhouse gases, especially carbon dioxide, is one of the main reasons. Countries have successively put forward carbon neutrality goals to mitigate the impact of climate change. However, a large amount of CO 2 produced in industrial emissions, coal-fired power generation, and chemical production processes is difficult to directly reduce, and carbon capture, utilization, and storage technologies have gradually become an important means to solve this problem.

[0004] Traditional carbon capture technologies mainly include chemical absorption, physical adsorption, and membrane separation. However, these technologies still face the following problems in practical applications: Insufficient selectivity: The membrane separation technology has low selectivity for CO 2 and is prone to simultaneously capturing other gases such as O 2 , N 2 , resulting in low purity. High energy consumption: Existing capture technologies consume a large amount of energy during the adsorption and desorption processes, especially under low-concentration CO 2 conditions. Complex operation: In large-scale industrial applications, equipment maintenance is complex, and the operation efficiency is significantly affected by changes in environmental conditions such as temperature, pressure, and humidity. Lack of intelligent optimization: Most systems rely on manual operation or fixed parameter operation, lack dynamic regulation ability, and cannot adapt to complex and changeable industrial environments.

[0005] With the rapid development of artificial intelligence technology, its advantages in optimizing complex industrial processes are gradually emerging. By combining sensor technology, data analysis, and machine learning algorithms, artificial intelligence technology can monitor the system status in real time, predict equipment performance, and optimize operating parameters, thus significantly improving carbon capture efficiency. Especially in membrane separation technology, the application of artificial intelligence technology can achieve: dynamically regulating the operating state of membrane materials, comprehensively analyzing historical data and real-time data to optimize capture strategies, reducing manual intervention, and improving the autonomy and stability of the system. In addition, in recent years, mixed matrix membranes, which are composite functional membrane materials formed by combining polymer organic matrices and inorganic fillers, and responsive materials have received extensive attention due to their adjustable and selective properties. Under the action of external stimuli, such materials can change their microscopic structures such as pore sizes, providing the possibility for the development of highly selective and low-energy-consuming carbon capture membranes. The combination of this material property and intelligent optimization control will make membrane separation technology move towards intelligence. Aiming at the above existing problems, the present invention aims to integrate intelligent optimization control and dynamically responsive membrane materials to solve the deficiencies of traditional carbon capture technology. Summary of the Invention

[0006] To solve the problems of insufficient selectivity of membranes, high system energy consumption, complex operation, and lack of intelligent optimization in traditional carbon capture technology, the present invention proposes a carbon capture system based on artificial intelligence optimization control.

[0007] The present invention provides a carbon capture system based on artificial intelligence optimization control, including a flue gas pipeline, a carbon capture membrane, a sensing module, a control module, and an execution module. A plurality of the carbon capture membranes are arranged in the flue gas pipeline along the radial direction of the flue gas pipeline. The pore size, pore size distribution, porosity, permeability, and thickness of the carbon capture membrane can all be adjusted. A sensing module is arranged in front of each carbon capture membrane. The sensing module is used to collect the operation data in the flue gas pipeline. The sensing module is connected to the input end of the control module and transmits the operation data in the flue gas pipeline to the control module. The output end of the control module is connected to the execution module. The control module is used to analyze and process the operation data to obtain an adjustment instruction and send the adjustment instruction to the execution module. The execution module is arranged in the flue gas pipeline. The execution module adjusts the gas temperature, humidity, and pressure in the flue gas pipeline through the adjustment instruction.

[0008] A carbon capture system based on artificial intelligence optimized control according to some embodiments of the present application. The carbon capture membrane includes a polymer organic matrix, inorganic fillers, and responsive materials. The polymer organic matrix includes one or more of a thermosensitive polymer, an intelligent polymer, and polyaniline. The inorganic fillers include one or more of ZIF-64, MLF-53, graphene, and graphene oxide. The responsive materials include one or more of a metal-organic framework MOF and an intelligent polymer.

[0009] A carbon capture system based on artificial intelligence optimized control according to some embodiments of the present application. A pretreatment device is provided at the inlet of the flue gas pipeline. The pretreatment device is used to remove solid impurities in the flue gas. The solid impurities include dust and particulate matter.

[0010] A carbon capture system based on artificial intelligence optimized control according to some embodiments of the present application. The sensing module includes a temperature sensor, a humidity sensor, a pressure sensor, and a gas concentration detector. The temperature sensor is used to detect the temperature in front of the carbon capture membrane. The humidity sensor is used to detect the humidity in front of the carbon capture membrane. The pressure sensor is used to detect the pressure in front of the carbon capture membrane. The gas concentration detector is used to detect the gas concentration in front of the carbon capture membrane. The operation data includes temperature data, pressure data, and gas concentration data.

[0011] A carbon capture system based on artificial intelligence optimized control according to some embodiments of the present application. The control module includes a data acquisition layer, a data preprocessing layer, a feature extraction and analysis layer, an adaptive regulation layer, a control decision layer, and a data output and feedback layer;

[0012] The data acquisition layer is used to receive the operation data, collect membrane performance data, and historical operation data of the carbon capture system, and transmit the operation data, membrane performance data, and historical operation data to the data preprocessing layer;

[0013] The data preprocessing layer is used to clean, denoise, and standardize the operation data, membrane performance data, and historical operation data to obtain a data set, and transmit the data set to the feature extraction and analysis layer;

[0014] The feature extraction and analysis layer is used to extract parameters related to the performance regulation of the carbon capture membrane from the data set, and transmit the parameters to the adaptive regulation layer;

[0015] The adaptive regulation layer is used to predict the future performance of the carbon capture membrane based on the parameters, plan and predict the operation parameters of the execution module based on the predicted future performance, and transmit the plan and prediction to the control decision layer;

[0016] The control decision-making layer is used to obtain an adjustment instruction based on the planning prediction and the membrane performance data, and transmit the adjustment instruction to the data output and feedback layer;

[0017] The data output and feedback layer transmits the adjustment instruction to the execution module and feeds back the adjustment instruction to the data acquisition layer.

[0018] According to some embodiments of the present application, a carbon capture system based on artificial intelligence optimization control, the adaptive regulation layer predicts the future performance of the carbon capture membrane through a deep learning model, and the deep learning model is one of LSTM or GRU.

[0019] According to some embodiments of the present application, a carbon capture system based on artificial intelligence optimization control, the control module further includes a real-time performance evaluation module, the real-time performance evaluation module performs real-time evaluation on the overall performance of the carbon capture system and obtains an evaluation result, and the overall performance of the carbon capture system includes carbon capture efficiency, the health state of the membrane and the stability of the system. The real-time performance evaluation module transmits the evaluation result to the adaptive regulation layer, and the adaptive regulation layer dynamically optimizes and adjusts the planning prediction through the evaluation result.

[0020] According to some embodiments of the present application, a carbon capture system based on artificial intelligence optimization control, the execution module includes a heating device, a cooling device, a drying device and a humidifying device. The heating device is used to receive the adjustment instruction to heat the gas in the flue gas pipeline, the cooling device is used to receive the adjustment instruction to cool the gas in the flue gas pipeline, the drying device is used to receive the adjustment instruction to dry the gas in the flue gas pipeline, and the humidifying device is used to receive the adjustment instruction to humidify the gas in the flue gas pipeline.

[0021] According to some embodiments of the present application, a carbon capture system based on artificial intelligence optimization control, a compressor is provided in the flue gas pipeline in front of each sensing module, and each compressor is connected to the control module. The compressor is used to receive the adjustment instruction to compress the gas in the flue gas pipeline and increase the pressure of the gas in the flue gas pipeline.

[0022] According to some embodiments of the present application, a carbon capture system based on artificial intelligence optimization control further includes a display device, the display device is connected to the control module, and the display device is used to display the real-time state, operation data and prediction analysis result of the carbon capture system.

[0023] A carbon capture system based on artificial intelligence optimization control proposed by the present invention integrates intelligent materials, dynamic regulation technology and artificial intelligence algorithms, thereby significantly improving CO2 The capture efficiency and selectivity are improved, and the energy consumption of the system operation is effectively reduced. The real-time monitoring of the performance of the carbon capture membrane is also realized, ensuring its adaptability and stability under various environmental conditions. On the one hand, the control module uses advanced data analysis technology to optimize the operation process and improve the overall operation efficiency of the system, enabling the system to maintain the best performance state for a long time. On the other hand, it uses artificial intelligence algorithms to optimize the working state of the membrane, accurately adjusting parameters such as the pore size and pore size of the carbon capture membrane according to real-time data, and achieving efficient selective capture of CO 2 , effectively avoiding the simultaneous capture of other gases. In addition, through intelligent optimization control, the energy consumption during the membrane regeneration process can be reduced, and real-time adjustment and optimization can further improve the stability and efficiency of the system operation. This carbon capture system is applicable to coal-fired power plants, chemical plants, thermal power plants and other high-emission industrial fields, with broad application value, providing important technical support and broad commercial application prospects for solving the global carbon emission problem. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 FIG. is a schematic structural diagram of a carbon capture system based on artificial intelligence optimization control according to an embodiment of the present invention;

[0025] Figure 2 FIG. is a schematic diagram of the working principle of the carbon capture membrane according to an embodiment of the present invention;

[0026] Figure 3 FIG. is a broken line graph of the number of experiments and errors of the control module according to an embodiment of the present invention;

[0027] Figure 4 FIG. is a schematic diagram of the test results of the performance error of the control module according to an embodiment of the present invention.

[0028] In the figure, 1, carbon capture membrane; 2, control module; 3, pretreatment equipment; 4, compressor. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0029] The following further describes in detail the embodiments of the present invention with reference to the drawings and embodiments. The following embodiments are used to illustrate the present invention, but cannot be used to limit the scope of the present invention.

[0030] Embodiment: This embodiment provides a carbon capture system based on artificial intelligence optimization control, as Figure 1As shown in the figure, it includes a flue gas pipeline, a carbon capture membrane 1, a sensing module, a control module 2, and an execution module. There are multiple carbon capture membranes 1 arranged along the radial direction of the flue gas pipeline in the flue gas pipeline. The pore size, pore size distribution, porosity, permeability, and thickness of the carbon capture membrane 1 can all be adjusted. In front of each carbon capture membrane 1, there is a sensing module, which is used to collect the operating data in the flue gas pipeline. The sensing module is connected to the input end of the control module 2 and transmits the operating data in the flue gas pipeline to the control module 2. The output end of the control module 2 is connected to the execution module. The control module 2 is used to analyze and process the operating data to obtain an adjustment instruction and send the adjustment instruction to the execution module. The execution module is arranged in the flue gas pipeline, and the execution module adjusts the gas temperature, humidity, and pressure in the flue gas pipeline through the adjustment instruction.

[0031] As a preference of this embodiment, specifically, the carbon capture membrane 1 includes a polymer organic matrix, inorganic fillers, and responsive materials. The polymer organic matrix includes one or more of thermosensitive polymers, intelligent polymers, and polyaniline. The inorganic fillers include one or more of ZIF-64, MLF-53, graphene, and graphene oxide. The responsive materials include one or more of metal-organic frameworks MOF and intelligent polymers. The pore size range of the carbon capture membrane 1 can be adjusted to to to adapt to the specific size of CO 2 molecules. The membrane pore size, pore size, pore size distribution, porosity, permeability, and thickness of the carbon capture membrane 1 can all change dynamically under external stimuli. The carbon capture membrane 1 adjusts the pore size through external temperature, pressure, surrounding gas concentration, especially CO 2 concentration and other signals, so as to achieve the selective capture of CO 2 . The carbon capture membrane 1 also has a regeneration design. The carbon capture membrane 1 can undergo a low-energy desorption process, and can recover pure CO 2 while maintaining a high capture efficiency. As Figure 2 shown, the working principle of the carbon capture membrane 1 is based on the selective permeability of the membrane material to different gas components. The carbon capture membrane 1 is a selective separation barrier that allows certain gases such as CO 2 to pass through preferentially, while other gases such as N 2 , O 2 are blocked. CO 2 can pass through the membrane faster due to its higher permeability and diffusion rate. And in this embodiment, the material of the carbon capture membrane 1 is selected as a CO 2 -philic material, which has excellent selective permeability performance for CO 2 . Specifically, the carbon capture membrane 1 relies on the difference in solubility or diffusivity of CO 2 on both sides of the separation membrane to capture CO 2 . When the gas passes through the membrane, CO 2The molecules will preferentially permeate through the membrane, thereby enriching CO on the permeate side downstream of the carbon capture membrane 1. 2 , while other gases remain in the upstream, that is, the return side, so as to achieve CO 2 Separation from other gases to maximize CO 2 The gas is accurately captured, thereby improving the efficiency of carbon capture. In addition, the surface of the carbon capture membrane 1 can also be coated with an anti-pollution coating, which can effectively reduce the pollution of particulate matter or other gases and extend the service life of the carbon capture membrane 1.

[0032] As a preference of this embodiment, specifically, a pre-treatment device 3 is provided at the entrance of the flue gas duct, and the pre-treatment device 3 is used to remove solid impurities in the flue gas, and the solid impurities include dust and particulate matter. As a preference of this embodiment, specifically, the sensing module includes a temperature sensor, a humidity sensor, a pressure sensor and a gas concentration detector, and the temperature sensor, the pressure sensor and the gas concentration detector are all arranged in the flue gas duct, the temperature sensor is used to detect the temperature in front of the carbon capture membrane 1, the humidity sensor is used to detect the humidity in front of the carbon capture membrane 1, the pressure sensor is used to detect the pressure in front of the carbon capture membrane 1, and the gas concentration detector is used to detect the gas concentration in front of the carbon capture membrane 1; the operation data includes temperature data, pressure data and gas concentration data. The temperature sensor, the pressure sensor and the gas concentration detector are all used to monitor the operation status in the pipeline in real time.

[0033] As a preferred embodiment of the present invention, the control module 2 includes a data acquisition layer, a data preprocessing layer, a feature extraction and analysis layer, an adaptive control layer, a control decision layer and a data output and feedback layer; the data acquisition layer is used to receive operation data and collect membrane performance data, historical operation data of the carbon capture system, and transmit the operation data, membrane performance data and historical operation data to the data preprocessing layer; the data preprocessing layer is used to clean, denoise and standardize the operation data, membrane performance data and historical operation data to obtain a data set, and transmit the data set to the feature extraction and analysis layer; the feature extraction and analysis layer is used to extract parameters related to the performance adjustment of the carbon capture membrane 1 from the data set, and transmit the parameters to the adaptive control layer, the parameters including CO 2Concentration change trends, membrane flux fluctuations, etc.; the adaptive regulation layer is used to predict the future performance of the carbon capture membrane 1 based on parameters, plan and predict the operating parameters of the execution module based on the predicted future performance, and transmit the planned prediction to the control decision-making layer. More preferably, the adaptive regulation layer predicts the future performance of the carbon capture membrane 1 through a deep learning model, and the deep learning model is one of LSTM or GRU. The adaptive regulation layer can also be based on reinforcement learning RL, continuously interact with the environment, automatically optimize the control strategy, adjust the operating parameters of the carbon capture membrane 1 to improve the carbon capture efficiency, use deep learning models such as LSTM or GRU to predict the future performance of the carbon capture membrane 1, and adjust the operating parameters in advance, and optimize the control strategy based on the prediction of the carbon capture system state to ensure that the operation of the predicted carbon capture membrane 1 is in the optimal working range; the control decision-making layer is used to obtain the adjustment instruction according to the planned prediction and the membrane performance data, and transmit the adjustment instruction to the data output and feedback layer. Through the adjustment instruction, the working parameters of the carbon capture membrane 1 can be adjusted in real time, such as the temperature, pressure, flow rate, gas composition, etc. of the carbon capture membrane 1, to ensure the stability and efficiency of the carbon capture process of the carbon capture membrane 1; the data output and feedback layer transmits the adjustment instruction to the execution module and feeds back the adjustment instruction to the data acquisition layer. The control module 2 comprehensively analyzes the historical data and real-time data, and intelligently adjusts the working parameters of the carbon capture membrane 1, such as the pore size, size, and distribution, etc., so that the carbon capture membrane 1 maintains the best capture efficiency under different working conditions, prevents the carbon capture membrane 1 from overloading or failing, and ensures the maximization of the capture efficiency. During the long-term use of the carbon capture membrane 1, its performance may decay. The control module 2 can automatically adjust the operation mode of the carbon capture system according to the decreasing trend of the capture efficiency, slow down the aging process of the carbon capture membrane 1, predict the service life of the carbon capture membrane 1 material or adsorbent, and give an early warning of the replacement time in advance. Based on the historical data, predict the possible failures of the carbon capture membrane 1 and give an early warning in advance, so as to reduce the decrease in carbon capture efficiency caused by equipment failures. In addition, the data output and feedback layer can also generate a detailed data analysis report to help users understand the performance change trend of the carbon capture membrane 1 and optimize the membrane use and maintenance strategy

[0034] In this embodiment, for different working conditions, such as high CO 2 concentration, low-temperature operation, etc., the carbon capture system adaptively selects the capture strategy: the control module 2 detects CO 2 concentration is relatively high, such as 40%, and automatically reduces the pore size of the carbon capture membrane 1 to to improve selectivity. At the same time, optimize the temperature and pressure to accelerate the capture rate. The control module 2 detects that the CO 2 concentration decreases, such as 10%, and automatically increases the pore size of the carbon capture membrane 1 to and at the same time adjusts the electric field to reduce energy consumption.

[0035] Preferably, in this embodiment, specifically, the control module 2 further includes a real-time performance evaluation module. The real-time performance evaluation module evaluates the overall performance of the carbon capture system in real time and obtains an evaluation result. The overall performance of the carbon capture system includes carbon capture efficiency, the health state of the membrane, and the stability of the carbon capture system. The real-time performance evaluation module transmits the evaluation result to the adaptive regulation layer. The adaptive regulation layer dynamically optimizes and adjusts the planning prediction based on the evaluation result, analyzes the energy consumption of the carbon capture membrane 1 under different working conditions, and reduces the energy consumption through optimization and adjustment, thereby improving the economy and environmental benefits of the overall carbon capture system.

[0036] The main body of the control module 2 in this embodiment utilizes a neural network model. The core of the control module 2 lies in its ability to combine historical data with real-time data so that the control module 2 can dynamically adjust the working parameters of the carbon capture membrane 1, thereby enabling the performance of the carbon capture membrane 1 to reach the optimal state. Through the training of the neural network, the performance of the carbon capture membrane 1 under different conditions can be simulated and predicted. Using the iterative training mechanism of the neural network, the control module 2 can continuously learn from new data and gradually improve the accuracy of its prediction. This self-optimizing ability is crucial for dealing with various changes encountered in actual operations because it allows the carbon capture system to quickly adjust and provide optimal performance when facing new operating conditions or environmental changes. Real-time performance feedback is another key advantage of neural network training. In the application of the carbon capture membrane 1, this means that the carbon capture system can monitor the operating state of the membrane and adjust the operating parameters according to real-time data to ensure the efficiency and stability of the carbon capture process. This ability to monitor and adjust in real time is of great significance for improving the reliability of the entire carbon capture system and reducing operating costs. In addition, the scalability of this algorithm system provides room for future technological progress. Figure 3 It shows an error trend varying with the number of experiments. The horizontal axis represents the number of experiments, and the vertical axis represents the error value. As the number of experiments increases, the error value fluctuates, but the error is not large. The fluctuations in the figure are related to changes in experimental conditions, such as the characteristics of the membrane material, operating temperature, and pressure. This indicates that the model has a high prediction accuracy under specific experimental conditions, while there is room for optimization under certain conditions. In practical applications, the control module 2 can adjust the model parameters by continuously learning experimental data, such as indicators like gas permeability and selectivity, thereby reducing the prediction error. Figure 4 It can be seen that after combining the neural network model with historical data and experimental data, the error is very small, only 0.014042, greatly improving the performance of the carbon capture membrane 1 and maximizing the carbon capture efficiency. Figure 3 and Figure 4It is the result obtained by combining the historical data of a certain power plant in the past five years, such as temperature, pressure, gas flow rate, etc., with the real-time data collected by sensors and inputting them into the control module 2 for analysis and optimization. It is composed of Figure 3 and Figure 4 It can be seen that the result predicted by the control module 2 is not much different from the real result, which shows the feasibility of this method. With the discovery of new materials and the development of new processes, the algorithm can be trained to incorporate this new information, thereby further improving the performance of the carbon capture membrane 1. This interdisciplinary integration involves not only materials science and chemical engineering, but also artificial intelligence and data analysis, providing a comprehensive and powerful tool for the development of carbon capture technology. By integrating these technologies into the algorithm system of the carbon capture membrane 1, it is not only possible to improve the efficiency of carbon capture, but also to provide an innovative solution for addressing global climate change and reducing greenhouse gas emissions. The implementation of this method is expected to play an important role in reducing dependence on fossil fuels, promoting the development of clean energy technologies, and achieving sustainable development goals.

[0037] As a preference of this embodiment, specifically, the execution module includes a heating device, a cooling device, a drying device, and a humidifying device. The heating device is used to receive an adjustment instruction to heat the gas in the flue gas pipeline, and the cooling device is used to receive an adjustment instruction to cool the gas in the flue gas pipeline. Through the heating device or the cooling device, the carbon capture system can adjust the operating temperature of the carbon capture membrane 1 to adapt to different climates or working conditions. During the long-term use of the carbon capture membrane 1, its performance may decay. The carbon capture system can automatically adjust the system operation mode according to the decreasing trend of the capture efficiency to slow down the aging process of the carbon capture membrane 1. The drying device is used to receive an adjustment instruction to dry the gas in the flue gas pipeline, and the humidifying device is used to receive an adjustment instruction to humidify the gas in the flue gas pipeline.

[0038] As a preference of this embodiment, specifically, a compressor 4 is provided in the flue gas pipeline in front of each sensing module, and each compressor 4 is connected to the control module 2. The compressor 4 is used to receive an adjustment instruction to compress the gas in the flue gas pipeline and increase the pressure of the gas in the flue gas pipeline.

[0039] Preferably, as an embodiment of the present invention, a display device is further included, which is connected to the control module 2. The display device is used to display the real-time status, operation data and prediction analysis results of the carbon capture system. Users can monitor important indicators such as the working condition of the membrane, the capture effect, and the energy consumption in real time through this interface. In addition, the carbon capture system of this embodiment can also be connected to the network. Users can view the working status of the carbon capture membrane 1 and the operation of the carbon capture system in real time through the network platform. If abnormalities occur, such as membrane blockage, unstable flow, etc., the carbon capture system will automatically issue an alarm and provide maintenance suggestions. The carbon capture system of this embodiment can also be connected to the management control system platform, which can centrally manage all the operating states of the carbon capture membrane 1 system, provide an interface for manual intervention to ensure manual adjustment or maintenance can be carried out in case of abnormalities. At the same time, the management control system platform can be docked with external systems, such as integration with the environmental protection supervision platform and the energy management system.

[0040] The following is the working process if the carbon capture system is provided with three carbon capture membranes 1, as Figure 1 shown, including: The flue gas emitted from power plants, chemical plants or steel plants first enters the pretreatment equipment 3 to remove solid impurities such as dust and particulate matter in the flue gas. The pretreated flue gas enters the first-stage compressor 4. After compression, the pressure of the gas is increased, enabling CO 2 molecules to pass through the carbon capture membrane 1 to the greatest extent, while other gases such as N 2 etc. are retained. After compression, the flue gas enters the first-stage carbon capture membrane 1, which is connected to the control module 2. Before the gas enters the first-stage carbon capture membrane 1 for separation, the sensing module real-time collects parameters such as the temperature, pressure, humidity, flow velocity and flow rate of the flue gas, and transmits the data to the control module 2. The control module 2 conducts comprehensive analysis through historical data and real-time data, and optimizes the working parameters of the carbon capture membrane 1 according to the analysis results, and sends the adjustment instructions to the execution module. The execution module dynamically adjusts the working parameters of the first-stage carbon capture membrane 1, such as the size and distribution of the membrane pores, etc., to achieve efficient capture of CO 2 . On the permeate side of the first-stage carbon capture membrane 1, the gas rich in CO 2 enters the next-stage compressor 4. At the same time, the carbon capture system continuously optimizes the control strategy and makes self-adjustment according to the operating state and feedback data. After further compressing to increase the gas pressure, it is sent to the second-stage carbon capture membrane 1, and the separation performance of the second-stage carbon capture membrane 1 is optimized through the control module 2. The CO 2 content in the gas on the permeate side of the second-stage carbon capture membrane 1 is significantly increased, and at the same time, there is a small amount of other gases. In addition, the retained gas of the second-stage carbon capture membrane 1 still contains a certain amount of CO 2, it needs to be returned to the first-stage carbon capture membrane 1 for re-separation. Similarly, the sensing module collects data in real time and sends it to the control module 2. The control module 2 calculates and sends adjustment instructions in real time, and the execution module dynamically adjusts the working parameters of the second-stage carbon capture membrane 1. After being processed by the first-stage carbon capture membrane 1 and the second-stage carbon capture membrane 1, the concentration of CO 2 in the flue gas can be increased to more than 70%. The gas on the permeate side then enters the next-stage compressor 4, and after the pressure is increased, it is sent to the third-stage carbon capture membrane 1. The separation process of the third-stage carbon capture membrane 1 is similar to the previous two stages. The retained gas of the third-stage carbon capture membrane 1 still contains a certain amount of CO 2 , so it needs to be returned to the second-stage carbon capture membrane 1 for re-separation. After three-stage separation, the concentration of CO 2 in the captured gas can be increased to more than 95%, meeting the subsequent utilization or storage requirements. All operation data and optimization suggestions are visually displayed through the management platform, facilitating real-time monitoring and scientific decision-making by users. Through this architecture, the carbon capture system can not only automatically optimize the working state of the membrane in a real-time environment, but also continuously learn and self-optimize to adapt to changing carbon capture requirements and environmental conditions. The carbon capture system supports the collection and analysis of large-scale industrial data, and can generate carbon capture efficiency reports and long-term operation optimization suggestions. Each module in the carbon capture system can work in coordination, and when capturing CO 2 , a process design of multi-stage and multi-segment carbon capture membranes 1 is adopted to maintain the ability of high selectivity, high efficiency and low energy consumption to capture CO 2 under dynamic conditions, providing technical support for industrial emission reduction and carbon neutrality goals. After capturing saturation, the adsorbed CO 2 in the carbon capture membrane 1 is released by heating or changing the electric field. The desorbed CO 2 is stored through a recovery device to achieve recycling. This embodiment can optimize the chemical properties and physical structure of the carbon capture membrane 1 at the molecular level, achieve high-selectivity capture of CO 2 . The carbon capture membrane 1 material can exclude other gases such as nitrogen and methane while capturing CO 2 , improving the capture purity and subsequent treatment efficiency. The development of the carbon capture membrane 1 material takes into account both degradability and recyclability, reducing the secondary pollution of waste to the environment.

[0041] The following is experimental verification: Experiment 1: Test conditions: Simulated industrial waste gas: 70% N 2 , 20% CO 2 , 10% O2. Operating temperature range: 20 - 60 °C. Operating pressure range: 1 - 10 bar. Test results: The CO 2 capture efficiency reaches more than 95%. Compared with traditional fixed-aperture membranes, the capture selectivity is increased by 40%, and the energy consumption is reduced by about 20%.

[0042] Experiment 2: Industrial application case. a. Application scenario industry: flue gas treatment in coal-fired power plants. b. Exhaust gas composition: N 2 : 75%, CO 2 : 15%, H 2 O: 5%, trace pollutants: 5%. Actual operation and capture process: The carbon capture system adjusts the operating parameters of the membrane according to the real-time detected exhaust gas flow rate and composition. The captured CO 2 is recovered through a low-temperature desorption system, and the concentration and purity reach over 99%. Maintenance and management: The carbon capture system conducts a health check on the performance of the carbon capture membrane 1 every week to predict the regeneration cycle of the carbon capture membrane 1. When the material of the carbon capture membrane 1 is about to reach the service life, the carbon capture system issues a replacement prompt in advance. In actual applications, sensors collect data such as temperature, pressure, gas component concentration, and membrane operating status such as capture rate and saturation. The data is transmitted to the control module 2 for real-time analysis and model training. Effect evaluation: Capture efficiency: The annual CO 2 capture reaches 200,000 tons. Energy consumption comparison: Compared with the traditional amine absorption method, the energy consumption is reduced by 25%. Economic benefits: Reduce carbon emission tax expenditures and at the same time increase the economic value of the recovered CO 2 , such as for the preparation of methanol.

[0043] The carbon capture system of this embodiment adopts a modular architecture, which can flexibly expand or reduce the carbon capture scale according to actual needs, and can be applied to high-carbon emission industries such as power plants, steel plants, and cement plants, as well as scenarios such as natural gas purification. In addition, if the carbon capture membrane 1 of this embodiment is replaced, this system can also be used for the capture of other gases.

[0044] The embodiments of the present invention are given for purposes of illustration and description, and are not exhaustive or limit the present invention to the disclosed form. Many modifications and variations are obvious to those of ordinary skill in the art. The embodiments are chosen and described in order to better illustrate the principles of the present invention and its practical applications, and to enable those of ordinary skill in the art to understand the present invention and design various embodiments with various modifications suitable for specific purposes.

Claims

1. A carbon capture system based on artificial intelligence optimization control, characterized in that: The invention comprises a flue gas duct, a carbon capture membrane (1), a sensor module, a control module (2) and an execution module. The flue gas duct is provided with a plurality of carbon capture membranes (1) arranged along the radial direction of the flue gas duct. The pore size, pore size distribution, porosity, permeability and thickness of the carbon capture membrane (1) are adjustable. The sensor module is arranged in front of each carbon capture membrane (1). The sensor module is used to collect operating data in the flue gas duct. The sensor module is connected to the input end of the control module (2) and transmits the operating data in the flue gas duct to the control module (2). The output end of the control module (2) is connected to the execution module. The control module (2) is used to analyze and process the operating data to obtain an adjustment instruction and send the adjustment instruction to the execution module. The execution module is arranged in the flue gas duct and adjusts the gas temperature, humidity and pressure in the flue gas duct according to the adjustment instruction.

2. The carbon capture system based on artificial intelligence optimization control according to claim 1, characterized in that: The carbon capture membrane (1) comprises a polymer organic matrix, an inorganic filler and a responsive material, wherein the polymer organic matrix comprises one or more of a thermosensitive polymer, a smart polymer and polyaniline, the inorganic filler comprises one or more of ZIF-64, MLF-53, graphene and graphene oxide, and the responsive material comprises one or more of a metal organic framework MOF and a smart polymer.

3. The carbon capture system based on artificial intelligence optimization control according to claim 1, characterized in that: A pre-treatment device (3) is provided at the inlet of the flue gas duct, and the pre-treatment device (3) is used to remove solid impurities in the flue gas, wherein the solid impurities include dust and particulate matter.

4. The carbon capture system based on artificial intelligence optimization control according to claim 1, characterized in that: The sensing module comprises a temperature sensor, a humidity sensor, a pressure sensor and a gas concentration detector; the temperature sensor is used to detect the temperature in front of the carbon capture membrane (1); the humidity sensor is used to detect the humidity in front of the carbon capture membrane (1); the pressure sensor is used to detect the pressure in front of the carbon capture membrane (1); and the gas concentration detector is used to detect the gas concentration in front of the carbon capture membrane (1); the operating data comprises temperature data, pressure data and gas concentration data.

5. The carbon capture system based on artificial intelligence optimization control according to claim 1, characterized in that: The control module (2) comprises a data acquisition layer, a data preprocessing layer, a feature extraction and analysis layer, an adaptive control layer, a control decision layer and a data output and feedback layer; The data acquisition layer is used to receive the operation data and collect membrane performance data, historical operation data of the carbon capture system, and transmit the operation data, membrane performance data and historical operation data to the data preprocessing layer; The data preprocessing layer is used to clean, denoise and standardize the operation data, membrane performance data and historical operation data to obtain a data set, and transmit the data set to the feature extraction and analysis layer; The feature extraction and analysis layer is used to extract parameters related to the performance adjustment of the carbon capture membrane (1) from the data set, and transmit the parameters to the adaptive control layer; The adaptive control layer is used to predict the future performance of the carbon capture membrane (1) based on the parameters, and to plan and predict the operating parameters of the execution module based on the predicted future performance, and transmit the planning and prediction to the control decision layer; The control decision layer is used to obtain adjustment instructions according to the planning prediction and the membrane performance data, and transmit the adjustment instructions to the data output and feedback layer; The data output and feedback layer transmits the adjustment instruction to the execution module, and feeds back the adjustment instruction to the data collection layer.

6. The carbon capture system based on artificial intelligence optimization control according to claim 5, characterized in that: The adaptive control layer predicts the future performance of the carbon capture membrane (1) through a deep learning model, and the deep learning model is one of LSTM or GRU.

7. The carbon capture system based on artificial intelligence optimization control according to claim 5, characterized in that: The control module (2) also includes a real-time performance evaluation module, which performs real-time evaluation on the overall performance of the carbon capture system and obtains an evaluation result, wherein the overall performance of the carbon capture system includes carbon capture efficiency, membrane health status and system stability. The real-time performance evaluation module transmits the evaluation result to the adaptive control layer, and the adaptive control layer dynamically optimizes and adjusts the planning prediction according to the evaluation result.

8. The carbon capture system based on artificial intelligence optimization control according to claim 1, characterized in that: The execution module includes a heating device, a cooling device, a drying device and a humidifying device. The heating device is used to receive the adjustment instruction to heat the gas in the flue gas duct, the cooling device is used to receive the adjustment instruction to cool the gas in the flue gas duct, the drying device is used to receive the adjustment instruction to dry the gas in the flue gas duct, and the humidifying device is used to receive the adjustment instruction to humidify the gas in the flue gas duct.

9. The carbon capture system based on artificial intelligence optimization control according to claim 1, characterized in that: A compressor (4) is provided in the flue gas duct in front of each of the sensor modules, and each of the compressors (4) is connected to the control module (2). The compressor (4) is used to receive the adjustment instruction to compress the gas in the flue gas duct and increase the pressure of the gas in the flue gas duct.

10. A carbon capture system based on artificial intelligence optimization control according to any one of claims 1 to 9, characterized in that: It also comprises a display device, which is connected to the control module (2) and is used to display the real-time status, operation data and prediction analysis results of the carbon capture system.

Images