A method and apparatus for performance optimization of a chemical absorption carbon capture system under varying operating conditions.

Abstract

This invention discloses a method and apparatus for performance optimization of a chemical carbon capture system under varying operating conditions. The method includes: acquiring historical operating data of the chemical carbon capture system and selecting key operating parameters; acquiring a chemical carbon capture model based on the mechanistic model and a physical information neural network of the chemical carbon capture system, and constructing a multi-objective optimization problem accordingly; acquiring varying operating conditions, determining constraints for the multi-objective optimization problem based on these conditions, solving the multi-objective optimization problem based on historical operating data and constraints, and obtaining a Pareto front solution set containing optimal values ​​for each key operating parameter; performing a weighted evaluation of the Pareto front solution set to obtain the optimal solution; and adjusting the key operating parameters to values ​​within the optimal solution. This invention achieves accurate simulation of the chemical carbon capture system by combining a mechanistic model and a physical information neural network, and realizes multi-objective collaborative optimization through the design of constraints under various varying operating conditions.

Description

Technical Field

[0001] This invention relates to a method and apparatus for performance optimization of a chemical carbon capture system under varying operating conditions, belonging to the technical field of performance optimization of chemical carbon capture systems. Background Technology

[0002] Among existing carbon capture and storage (CCS) technologies, such as pre-combustion capture, oxy-fuel combustion, and post-combustion capture, post-combustion carbon dioxide capture using amine solvents such as monoethanolamine (MEA) shows promising commercial application prospects due to its ease of integration into existing power plants, high carbon dioxide capture efficiency, and flexible operation under various conditions. However, the main drawback of this technology is that the solvent regeneration process consumes a large amount of steam, leading to a decrease in the overall efficiency of the power plant and thus significantly increasing capture costs.

[0003] In recent years, performance optimization under varying operating conditions has been considered a key technology for improving the operating efficiency of chemical carbon capture systems and achieving economic and low-carbon goals. Its importance is mainly reflected in the following two aspects: (1) Coal-fired power plants frequently need to participate in grid power regulation to cope with the dynamic balance between the supply and demand sides. With the growth of electricity demand and the widespread application of renewable energy sources such as wind and solar power, this need has become increasingly urgent. Coal-fired power plants must respond quickly to changes in operating conditions over a wide load range. Carbon capture systems with performance optimization for variable operating conditions can help power plants adapt to this flexible operating requirement and ensure the stability and efficiency of the system under different loads.

[0004] (2) High operating costs limit the large-scale deployment of afterburning capture technology in power plants. Therefore, operating the afterburning capture system at full load all the time is not an economically feasible option. Through variable operating condition performance optimization, the afterburning capture system can flexibly adjust its operating strategy according to external conditions (such as electricity price fluctuations). When electricity prices are high, steam used for carbon dioxide capture is reduced, and more steam is used for power generation to improve economic efficiency; while when electricity prices are low, steam extraction is increased to improve the carbon dioxide capture rate, thereby achieving the low-carbon goal.

[0005] In the current research field of carbon capture technology, exploration of performance optimization under varying operating conditions has gradually begun, but existing optimization methods still have certain limitations. Existing technologies mainly optimize under varying operating conditions in the following ways: Single model modeling: Calculating the optimal parameters under varying operating conditions based on pure mechanistic models (simulating according to physical laws) or pure data models (trained with historical data); Single objective optimization: Most studies focus on optimizing a single objective (such as minimizing energy consumption or maximizing capture rate).

[0006] The main shortcomings are as follows: it is difficult to balance model accuracy and generalization. Pure mechanistic models have poor adaptability to actual operating data, while pure data models lack physical constraints and have poor extrapolation, making it difficult to achieve a balance between model accuracy and generalization ability. There is a lack of multi-objective balance. Optimization of a single objective is prone to "losing sight of one thing while focusing on another" (such as reducing energy consumption leading to a failure to meet the capture rate standard). The synergy and trade-off between energy consumption, cost, and capture rate are not considered, thus limiting applicability. Summary of the Invention

[0007] The purpose of this invention is to provide a method and apparatus for performance optimization of a chemical carbon capture system under varying operating conditions, thereby solving the problems of difficulty in balancing model accuracy and generalization, and lack of multi-objective balance in the existing technology for performance optimization of chemical carbon capture systems under varying operating conditions.

[0008] To achieve the above objectives, the present invention employs the following technical solution: In a first aspect, the present invention provides a method for performance optimization of a chemical absorption carbon capture system under varying operating conditions, comprising: After obtaining historical operating data of the chemical absorption carbon capture system, the key operating parameters affecting the performance of the chemical absorption carbon capture system were screened out. Obtain the mechanism model of the chemical absorption carbon capture system and the chemical absorption carbon capture model constructed by the physical information neural network, and construct a multi-objective optimization problem based on the chemical absorption carbon capture model; Obtain variable operating conditions, determine the constraints of the multi-objective optimization problem based on the variable operating conditions, solve the multi-objective optimization problem based on historical operating data and constraints, and obtain the Pareto front solution set containing the optimal values ​​of each key operating parameter; The optimal solution is obtained by weighting the Pareto front solution set using the TOPSIS method. The key operating parameters of the chemical absorption carbon capture system are adjusted to the optimal values ​​to complete performance optimization.

[0009] Furthermore, after obtaining the historical operating data of the chemical carbon capture system, the key operating parameters affecting the performance of the chemical carbon capture system are screened out using the following formula: ; in, This represents a key data matrix composed of key operating parameters. This represents the original data matrix composed of runtime parameters, with dimensions of [missing information]. , Indicates the number of samples. Indicates the number of runtime parameters. This represents the defined mean vector. This represents the set principal component loading matrix, which is obtained by eigenvalue decomposition of the data covariance matrix of the historical operating data. This indicates transpose.

[0010] Furthermore, the chemical absorption carbon capture model is constructed using the following method: Using the mechanism model of the chemical absorption carbon capture system as the core framework, a physical information neural network is introduced to correct the mechanism model, resulting in a chemical absorption carbon capture model. The aforementioned mechanism model is a mathematical model based on the principles of mass conservation, energy conservation, and chemical reaction equilibrium, describing the chemical absorption carbon capture system under stable operating conditions. The method of introducing a physical information neural network to correct the mechanism model includes: predicting the reaction rate constant affected by operating parameters through the physical information neural network, feeding back the predicted value of the reaction rate constant to the mechanism model, and realizing parameter updates and state correction of the mechanism model under varying operating conditions.

[0011] Furthermore, the expression for the multi-objective optimization problem is: ; in, This represents the carbon dioxide capture rate. It is the total energy consumption of the chemical carbon capture process. This is the total operating cost of the chemical carbon capture system. Indicates the steam extraction flow rate. Indicates the reboiler temperature. Indicates regeneration pressure, Indicates flue gas flow rate, To maximize, This indicates minimization.

[0012] Furthermore, when the variable operating condition is fluctuating flue gas flow rate, the expression for the constraint condition is: ; in, This represents the carbon dioxide capture rate. Indicates flue gas flow rate, The nominal value representing the set flue gas flow rate. Indicates the steam extraction flow rate. This indicates the maximum set extraction steam flow rate. Indicates the reboiler temperature. This indicates regeneration pressure.

[0013] Furthermore, when the variable operating condition is that the steam extraction capacity is limited, the expression for the constraint condition is: ; in, This represents the carbon dioxide capture rate. Indicates the steam extraction flow rate. This indicates the maximum set extraction steam flow rate. Indicates the reboiler temperature. This indicates regeneration pressure.

[0014] Furthermore, when the variable operating condition is a capture rate constraint, the expression for the constraint is: ; in, This represents the carbon dioxide capture rate. Indicates the steam extraction flow rate. This indicates the maximum set extraction steam flow rate. Indicates the reboiler temperature. This indicates regeneration pressure.

[0015] Furthermore, when the variable operating condition is a production constraint, the expression for the constraint is: ; in, This represents the carbon dioxide capture rate. Indicates the steam extraction flow rate. This indicates the maximum set extraction steam flow rate. Indicates the reboiler temperature. Indicates regeneration pressure, Indicates carbon dioxide capture yield. This represents the set minimum carbon dioxide capture yield. This indicates the maximum set carbon dioxide capture yield; The carbon dioxide capture yield is calculated using the following formula: ; in, This represents the concentration of carbon dioxide in the flue gas. This indicates the flue gas flow rate.

[0016] Furthermore, the step of using the TOPSIS method to perform a weighted evaluation of the Pareto front solution set to obtain the optimal solution includes: If the Pareto front solution set contains Given several solutions, construct the decision matrix for each key operating parameter. The expression for the decision matrix is ​​as follows: ; in, Represents the decision matrix. Indicates the first The carbon dioxide capture rate of each solution. Indicates the first The total energy consumption of the chemical absorption carbon capture process for each solution. Indicates the first The total operating cost of a chemical absorption carbon capture system for each solution; The elements in the decision matrix are normalized using the following formula: ; =1,2…, ; =1,2,3; in, express The normalized value, For the first The solution of the first... One goal, when When = 1 corresponds to the carbon dioxide capture rate, when When =2 corresponds to the total energy consumption of the chemical carbon capture process, when The total operating cost of the chemical absorption carbon capture system when the value is 3; The normalized values ​​of each element in the decision matrix are weighted and normalized using the following formula: ; in, express The weighted normalized value, For the first The weight of each objective; The positive and negative ideal solutions are calculated based on the weighted normalized values ​​using the following formula: ; ; in, This represents the positive ideal solution for the carbon dioxide capture rate. express The weighted normalized value, Indicates the first The carbon dioxide capture rate of each solution. The positive ideal solution representing the total energy consumption of the chemical carbon capture process is... Indicates the first The total energy consumption of the chemical absorption carbon capture process for each solution. The positive ideal solution represents the total operating cost of a chemical absorption carbon capture system. Indicates the first The total operating cost of a chemical absorption carbon capture system for a given solution. The negative ideal solution represents the capture rate of carbon dioxide. The negative ideal solution represents the total energy consumption of the chemical carbon capture process. The negative ideal solution represents the total operating cost of a chemical absorption carbon capture system. This indicates taking the maximum value. This indicates taking the minimum value; The distance from each solution in the Pareto front solution set to the positive ideal solution is calculated using the following formula: ; in, Indicates the first The distance from each solution to the ideal solution. Indicates the first The positive ideal solution for each objective; The distance from each solution in the Pareto front solution set to the negative ideal solution is calculated using the following formula: ; in, Indicates the first The distance from the ideal solution to the negative ideal solution Indicates the first The negative ideal solution for each objective; The overall score is calculated based on the distances from each solution in the Pareto front solution set to the positive and negative ideal solutions, using the following formula: ; in, Indicates the first The overall score of each solution; choose The largest solution is the optimal solution.

[0017] Secondly, the present invention provides a performance optimization device for a chemical absorption carbon capture system under varying operating conditions, comprising: The key operating parameter screening module is configured to: after acquiring historical operating data of the chemical carbon capture system, screen out the key operating parameters that affect the performance of the chemical carbon capture system; The multi-objective optimization problem construction module is configured to: obtain a chemical carbon capture model based on the mechanism model and physical information neural network of the chemical carbon capture system, and construct a multi-objective optimization problem based on the chemical carbon capture model; The multi-objective optimization problem solving module is configured to: acquire variable operating conditions, determine the constraints of the multi-objective optimization problem based on the variable operating conditions, solve the multi-objective optimization problem based on historical operating data and constraints, and obtain a Pareto front solution set containing the optimal values ​​of each key operating parameter; The optimal solution selection module is configured to use the TOPSIS method to perform a weighted evaluation of the Pareto front solution set to obtain the optimal solution. The performance optimization module is configured to adjust the key operating parameters of the chemical absorption carbon capture system to the optimal values ​​to complete the performance optimization.

[0018] Compared with the prior art, the beneficial effects achieved by the present invention are: This invention provides a performance optimization method and apparatus for a chemical carbon capture system under varying operating conditions. The chemical carbon capture model is constructed using a mechanistic model and a physical information neural network. The mechanistic model provides the physical foundation, while the physical information neural network integrates physics and data, enabling the model to possess both the reliability of the mechanistic model and the ability to accurately describe the varying operating conditions. Together, they achieve precise simulation of the performance of the chemical carbon capture system, thus enabling its wide application in system optimization and scheduling under different varying operating conditions. This greatly enhances the practicality and versatility of the method, effectively meeting the scheduling needs under various complex operating conditions. By comprehensively considering multiple varying operating conditions, multi-objective collaborative optimization is achieved, and the TOPSIS method is applied to perform subsequent decision analysis on the obtained Pareto front solution set to determine whether it meets the requirements of the actual process. These two parts effectively solve the shortcomings of existing technologies in optimizing the performance of chemical carbon capture systems under varying operating conditions, such as inaccurate modeling and insufficient multi-objective collaboration, improving the reliability and practicality of optimizing chemical carbon capture systems under varying operating conditions. Attached Figure Description

[0019] Figure 1 This is a flowchart of the performance optimization method for the chemical absorption carbon capture system under varying operating conditions provided in Example 1; Figure 2 This is a flowchart of the performance optimization method for the chemical absorption carbon capture system under various operating conditions provided in Example 2. Detailed Implementation

[0020] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to illustrate the technical solution of the present invention more clearly, and should not be used to limit the scope of protection of the present invention.

[0021] Example 1

[0022] like Figure 1 As shown, this embodiment provides a performance optimization method for a chemical absorption carbon capture system under varying operating conditions, including: After obtaining historical operating data of the chemical absorption carbon capture system, the key operating parameters affecting the performance of the chemical absorption carbon capture system were screened out. Obtain the mechanism model of the chemical absorption carbon capture system and the chemical absorption carbon capture model constructed by the physical information neural network, and construct a multi-objective optimization problem based on the chemical absorption carbon capture model; Obtain variable operating conditions, determine the constraints of the multi-objective optimization problem based on the variable operating conditions, solve the multi-objective optimization problem based on historical operating data and constraints, and obtain the Pareto front solution set containing the optimal values ​​of each key operating parameter; The optimal solution is obtained by weighting the Pareto front solution set using the TOPSIS method. The key operating parameters of the chemical absorption carbon capture system are adjusted to the optimal values ​​to complete performance optimization.

[0023] This invention constructs a chemical carbon capture model using a mechanistic model and a physical information neural network. The mechanistic model provides the physical foundation, while the physical information neural network integrates physics and data, enabling the model to possess both the reliability of the mechanistic model and the ability to accurately describe processes under varying operating conditions. Together, they achieve precise simulation of the performance of the chemical carbon capture system, thus allowing for wide application in system optimization and scheduling under different varying operating conditions. This greatly enhances the practicality and versatility of the method and can effectively meet the scheduling needs under various complex operating conditions. By comprehensively considering multiple varying operating conditions, multi-objective collaborative optimization is achieved, and the TOPSIS method is applied to perform subsequent decision analysis on the obtained Pareto front solution set to determine whether it meets the requirements of the actual process. Through the above two parts of the design, the shortcomings of existing technologies in optimizing the performance of chemical carbon capture systems under varying operating conditions, such as inaccurate modeling and insufficient multi-objective collaboration, are effectively solved, improving the reliability and practicality of optimizing chemical carbon capture systems under varying operating conditions.

[0024] Example 2

[0025] The present invention will now be described in detail with reference to the accompanying drawings.

[0026] To achieve economic efficiency and low-carbon operation of chemical carbon capture systems, it is essential to develop a performance optimization method for variable operating conditions. This method needs to comprehensively consider the system's energy consumption, cost, and capture efficiency under different operating loads to achieve overall optimized scheduling and rapid response capabilities.

[0027] like Figure 2 As shown in the example, this paper provides a performance optimization method for a chemical absorption carbon capture system under varying operating conditions. The steps are as follows: Step S1: Determine key operating parameters.

[0028] Based on historical plant operating data or laboratory measurement data, analyze the key operating parameters that affect the performance of the chemical absorption carbon capture system under varying operating conditions (such as flue gas flow fluctuations, limited steam extraction, capture rate or production constraints).

[0029] Step S2: Construct a chemical absorption carbon capture system model that combines mechanistic modeling with data-driven approaches.

[0030] The mechanism model describes the core chemical reaction and mass and heat transfer processes. Combined with the Physics-Informed Neural Networks (PINN), it uses actual operating data to adjust the reaction rate constant affected by operating parameters, thereby achieving accurate prediction of the reaction rate constant.

[0031] This step aims to establish a chemical carbon capture model capable of adapting to varying operating conditions. The model uses a mechanistic model as its core framework and incorporates a Physical-Informed Neural Network (PINN) for calibration, together forming a chemical carbon capture model. Its core connections and construction method are as follows: Mechanism model: A mathematical model constructed based on physicochemical principles such as mass conservation, energy conservation, and chemical reaction equilibrium, describing the steady-state operation of a chemical carbon capture system. It is responsible for characterizing the core chemical reactions and mass and heat transfer processes in the chemical carbon capture system (such as absorption towers and regeneration towers), providing a steady-state benchmark with clear physical meaning for the system behavior.

[0032] PINN's main function is to accurately predict the reaction rate constant affected by operating parameters and feed this prediction back to the mechanistic model. During varying operating conditions, this enables parameter updates and state corrections for the mechanistic model. The relationship between PINN and the mechanistic model: PINN's loss function inherits the physical laws followed by the mechanistic model, ensuring that its data-driven predictions do not violate fundamental physical principles.

[0033] The connection between the two and the chemical absorption carbon capture model: the mechanistic model provides the physical basis, and PINN integrates physics and data. Together, they constitute a high-precision chemical absorption carbon capture system model, which not only has the reliability of the mechanistic model, but also has the ability to accurately describe the process under varying operating conditions, thus achieving accurate simulation of the performance of the chemical absorption system.

[0034] The Relationship Between Chemical Absorption Carbon Capture Models and Multi-Objective Optimization Problems: The chemical carbon capture model established in step S2 forms the computational basis for the multi-objective optimization (energy efficiency, economy, and environment) in step S3. In the S3 optimization, based on the chemical carbon capture model, the system performance (such as CO2 capture rate, energy consumption, and cost) under different operating parameters can be quickly calculated. Simultaneously, the physical constraints within the chemical carbon capture system model ensure that the predicted results conform to actual conditions.

[0035] Step S3: Establish a multi-objective optimization problem.

[0036] Under varying operating conditions (such as flue gas flow fluctuations, limited steam extraction, and constraints on capture rate or production), a multi-objective optimization problem with low carbon emissions and economic efficiency as objectives is established.

[0037] Step S4: Solve the multi-objective optimization problem.

[0038] The Pareto front solution set is obtained by solving multi-objective optimization problems using a genetic algorithm.

[0039] Step S5: Perform a weighted evaluation of the Pareto front solution set and select the optimal solution.

[0040] The TOPSIS (Technique for Order Preference by Similarity to an Ideal Solution) method is used to perform a weighted evaluation of the Pareto front solution set. Based on actual process requirements, the optimal solution is selected to achieve low-carbon economic goals and improve system performance. The TOPSIS method ranks a finite number of evaluation objects according to their proximity to the ideal target.

[0041] In step S1, under different operating conditions, it is first necessary to identify the key operating parameters affecting the performance of the chemical absorption carbon capture system. Key operating parameters, including the extraction steam flow rate, are identified using principal component analysis (PCA). Reboiler temperature flue gas flow rate and regeneration pressure The PCA calculation formula is shown in formula (1): (1); in, This represents the key data matrix, which consists of key operating parameters. This represents the original data matrix, composed of runtime parameters. The dimensions of the original data matrix are... , Indicates the number of samples. Indicates the number of runtime parameters. This represents the defined mean vector. This represents the principal component loading matrix, obtained by eigenvalue decomposition of the covariance matrix of historical operating data. This indicates transpose.

[0042] In step S2, based on the mechanistic model and combined with a Physical Information Neural Network (PINN), the core chemical reaction and mass and heat transfer processes are integrated, and the reaction rate constant affected by operating parameters is adjusted using actual operating data. The loss function of PINN is as follows: (2); in, Describes the loss function of PINN. The data loss function measures the difference between the model's predictions and the actual values, and is one of the objective functions that needs to be minimized during model training. The physics-based loss function incorporates physical laws to constrain the model, ensuring that the model's output conforms to specific physical principles. It is also one of the objectives to be minimized during model training. Weights By balancing data fit and physical compliance through hyperparameter tuning (such as grid search), the role of physical constraints can be strengthened when data is sparse.

[0043] The data loss function (fitting historical or experimental data) is shown in formula (3): (3); in, This indicates the number of samples in the actual operational dataset, i.e., the number of data points used to calculate the data loss function. These samples are actual operational data, derived from on-site monitoring of the chemical absorption carbon capture system, and are used for data-driven calibration of the reaction rate constant in the model parameters. The samples are not randomly generated, but rather real engineering data, ensuring the model's practicality under varying operating conditions. This represents the first index variable, used to iterate through each sample in the dataset, with values ​​ranging from 1 to... ; Represents the first in the dataset One sample, Expand to ,in For the first Operating parameters, such as flue gas flow rate and temperature measured on-site; Represents a parameter A defined function (usually a neural network model) that uses As input, the predicted value of the output reaction rate constant is mainly predicted to be affected by operating parameters (such as flue gas flow rate, temperature, and pressure). Representation function The set of parameters in a neural network model, Typically, a neural network model includes parameters such as a weight matrix and a bias vector. The goal of training a neural network model is to find an optimal set of parameters. Minimize the loss function; Represents the first in the dataset Sample The corresponding true value, that is, the correct result expected from the model output; This represents calculating the square of the Euclidean norm, used to calculate predicted values. and the true value The distance between them measures the magnitude of their difference.

[0044] The physical constraint loss function (differential equation residual) is shown in Equation (4): (4); in, This indicates the number of samples or constraints related to physical constraints. This represents the second index variable, used to iterate through samples or constraints related to physical constraints, with values ​​ranging from 1 to... ; Represents the first related to physical constraints An input quantity or state variable, which may be related to certain parameters or states of the chemical absorption carbon capture system; It is a residual function defined based on physical laws, which is... and parameters of neural network models As input, the output is a residual value that reflects the degree to which physical constraints are satisfied; This represents the square of the Euclidean norm, used to measure the residual function. The magnitude of the physical constraint is the degree to which it is not satisfied.

[0045] The core equations of the mechanistic model include: mass balance equation, reaction kinetics equation, energy balance equation, and gas-liquid equilibrium.

[0046] The mass balance equations include gas phase mass balance, liquid phase mass balance, and interfacial balance; the reaction kinetic equations, taking MEA (monoethanolamine) as an example, show that the reaction between carbon dioxide and MEA mainly includes carbonate formation and amino group reaction; the energy balance equations include gas phase energy balance and liquid phase energy balance; gas-liquid balance: the balance of carbon dioxide at the gas-liquid interface is described by Henry's law or empirical correlation.

[0047] The mechanistic model comprehensively considers mass balance, reaction kinetic equations, energy balance equations, and gas-liquid balance. The core equations describe the gas-liquid mass transfer, chemical reaction, and thermal effects. By using the least squares method and combining experimental data and historical plant operation data, the main model parameters are fitted to achieve accurate prediction of the model.

[0048] In step S3, the expression for the multi-objective optimization problem is as follows: (5); in, This represents the carbon dioxide capture rate. It is the total energy consumption of the chemical carbon capture process. This is the total operating cost of the chemical carbon capture system. Indicates the steam extraction flow rate. Indicates the reboiler temperature. This indicates regeneration pressure.

[0049] The constraints under different variable operating conditions are as follows: Variable operating condition 1: Flue gas flow rate fluctuation, the flue gas flow rate is within the nominal value of the flue gas flow rate (denoted as...). Flue gas flow rate fluctuates within ±10% of the range of flue gas flow rate. The range of values ​​is .

[0050] The constraints under the variable operating conditions are: (6); in, Indicates the steam extraction flow rate. This indicates the maximum set extraction steam flow rate. Indicates the reboiler temperature. This indicates regeneration pressure.

[0051] Variable operating condition 2: Limited steam extraction capacity reduces the steam supply to the regeneration tower, affecting the regeneration effect. The capture rate requirement is appropriately relaxed to 85%.

[0052] The constraints under variable operating condition two are: (7).

[0053] Variable operating condition three: Capture rate is limited. To meet stringent environmental regulations, the carbon capture rate must reach at least 95%, i.e. .

[0054] The constraints under variable operating condition three are: (8).

[0055] Variable operating condition four: Production constraint. To meet the demand for carbon dioxide production from downstream processes or the market, the chemical absorption carbon capture system must ensure a certain carbon dioxide capture output. Within a specific range, that is . This represents the set minimum carbon dioxide capture yield. This indicates the maximum set carbon dioxide capture yield.

[0056] The formula for calculating carbon dioxide capture yield is as follows: ,in, This represents the concentration of carbon dioxide in the flue gas. This indicates the flue gas flow rate.

[0057] The constraints under variable operating condition four are: (9).

[0058] In step S4, the non-dominated sorting genetic algorithm (NSGA-II) is used to solve the multi-objective optimization problem constructed in step S3, generating a Pareto front solution set. The expression is: (10); Each solution Corresponding to a set of performance indicators , This indicates the preferred value for the extraction steam flow rate. This indicates the preferred value for the reboiler temperature. This indicates the preferred value for regeneration pressure. The preferred value representing the carbon dioxide capture rate. This represents the preferred value for the total energy consumption of the chemical carbon capture process. This represents the preferred value for the total operating cost of a chemical absorption carbon capture system.

[0059] In step S5, the TOPSIS method is used to perform a weighted evaluation of the Pareto front solution set and select the optimal solution. The specific steps are as follows: (1) Constructing the decision matrix: Assume that the Pareto front solution set contains One solution, three objectives ( , , The decision matrix is ​​as follows: (11); in, Represents the decision matrix. Indicates the first The carbon dioxide capture rate of each solution. Indicates the first The total energy consumption of the chemical absorption carbon capture process for each solution. Indicates the first The total operating cost of a chemical absorption carbon capture system.

[0060] (2) Normalization process: ; =1,2…, ; =1,2,3(12) in, express The normalized value, For the first The solution of the first... One goal, when When = 1 corresponds to the carbon dioxide capture rate, when When =2 corresponds to the total energy consumption of the chemical carbon capture process, when =3 corresponds to the total operating cost of the chemical absorption carbon capture system.

[0061] (3) Weighted normalization: (13); in, express The weighted normalized value, For the first The weight of each objective, .

[0062] (4) Determine the positive and negative ideal solutions: Positive ideal solution: (14); Negative ideal solution: (15); in, The positive ideal solution representing the capture rate of carbon dioxide (the positive ideal solution for the first objective). express The weighted normalized value, Indicates the first The first objective of the solution, i.e. the first... The carbon dioxide capture rate of each solution. The positive ideal solution representing the total energy consumption of the chemical carbon capture process is... Indicates the first The second objective of the solution, namely the first The total energy consumption of the chemical absorption carbon capture process for each solution. The positive ideal solution represents the total operating cost of a chemical absorption carbon capture system. Indicates the first The third objective of the solution, namely the first The total operating cost of a chemical absorption carbon capture system for a given solution. The negative ideal solution represents the capture rate of carbon dioxide. The negative ideal solution represents the total energy consumption of the chemical carbon capture process. The negative ideal solution represents the total operating cost of a chemical absorption carbon capture system. This indicates taking the maximum value. This indicates taking the minimum value.

[0063] (5) Calculate distance and relative proximity: No. The distance from each Pareto front solution to the positive ideal solution is calculated using the following formula: (16); in, Indicates the first The distance from each solution to the ideal solution. Indicates the first The ideal solution to a given objective.

[0064] No. The distance from each Pareto front solution to the negative ideal solution is calculated using the following formula: (17); in, Indicates the first The distance from the ideal solution to the negative ideal solution Indicates the first The negative ideal solution of an objective.

[0065] The overall score (relative closeness) is calculated using the following formula: (18); in, Indicates the first The overall score of each solution.

[0066] (6) Select the optimal solution: The largest solution is the optimal solution.

[0067] After obtaining the optimal solution, the three objectives of the chemical carbon capture system are adjusted to the three objective values ​​in the optimal solution, thus completing the performance optimization of the chemical carbon capture system under varying operating conditions.

[0068] Based on the above analysis, this patent addresses the shortcomings of the prior art through the following innovative points: 1. Establish a hybrid model that integrates mechanistic model and data-driven approach: The physical constraint loss function of PINN ensures that the model output conforms to physical laws, while the data loss function improves the model's fitting accuracy to actual operating data, thus combining the reliability of mechanistic model with the adaptability of data model.

[0069] 2. Multivariate Collaborative Optimization: This approach comprehensively considers various variable operating conditions such as flue gas flow fluctuations, limited steam extraction, and capture rate constraints to achieve collaborative optimization of multiple objectives, including energy consumption, cost, and capture rate. A genetic algorithm is used to solve the multi-objective optimization problem, obtaining the Pareto front solution set. The TOPSIS method is then applied to perform subsequent decision analysis on the obtained Pareto front solution set to determine whether it meets the requirements of the actual process.

[0070] Therefore, this invention effectively solves the defects of existing variable operating condition performance optimization technology, such as inaccurate process modeling and insufficient multi-objective coordination, and improves the reliability and practicality of variable operating condition optimization of carbon capture systems.

[0071] Example 3

[0072] This embodiment provides a performance optimization device for a chemical absorption carbon capture system under varying operating conditions, including: The key operating parameter screening module is configured to: after acquiring historical operating data of the chemical carbon capture system, screen out the key operating parameters that affect the performance of the chemical carbon capture system; The multi-objective optimization problem construction module is configured to: obtain a chemical carbon capture model based on the mechanism model and physical information neural network of the chemical carbon capture system, and construct a multi-objective optimization problem based on the chemical carbon capture model; The multi-objective optimization problem solving module is configured to: acquire variable operating conditions, determine the constraints of the multi-objective optimization problem based on the variable operating conditions, solve the multi-objective optimization problem based on historical operating data and constraints, and obtain a Pareto front solution set containing the optimal values ​​of each key operating parameter; The optimal solution selection module is configured to use the TOPSIS method to perform a weighted evaluation of the Pareto front solution set to obtain the optimal solution. The performance optimization module is configured to adjust the key operating parameters of the chemical absorption carbon capture system to the optimal values ​​to complete the performance optimization.

[0073] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0074] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0075] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0076] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0077] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for performance optimization of a chemical absorption carbon capture system under varying operating conditions, characterized in that, include: After obtaining historical operating data of the chemical absorption carbon capture system, the key operating parameters affecting the performance of the chemical absorption carbon capture system were screened out. Obtain the mechanism model of the chemical absorption carbon capture system and the chemical absorption carbon capture model constructed by the physical information neural network, and construct a multi-objective optimization problem based on the chemical absorption carbon capture model; Obtain variable operating conditions, determine the constraints of the multi-objective optimization problem based on the variable operating conditions, solve the multi-objective optimization problem based on historical operating data and constraints, and obtain the Pareto front solution set containing the optimal values ​​of each key operating parameter; The optimal solution is obtained by weighted evaluation of the Pareto front solution set using the TOPSIS method. The key operating parameters of the chemical absorption carbon capture system are adjusted to the optimal values ​​to achieve performance optimization. The chemical absorption carbon capture model was constructed using the following method: Using the mechanism model of the chemical absorption carbon capture system as the core framework, a physical information neural network is introduced to correct the mechanism model, resulting in a chemical absorption carbon capture model. The aforementioned mechanism model is a mathematical model based on the principles of mass conservation, energy conservation, and chemical reaction equilibrium, describing the chemical absorption carbon capture system under stable operating conditions. The method of introducing a physical information neural network to correct the mechanism model includes: predicting the reaction rate constant affected by operating parameters through the physical information neural network, feeding back the predicted value of the reaction rate constant to the mechanism model, and realizing parameter updates and state correction of the mechanism model under varying operating conditions. The expression for the multi-objective optimization problem is: ； in, This represents the carbon dioxide capture rate. It is the total energy consumption of the chemical carbon capture process. This is the total operating cost of the chemical carbon capture system. Indicates the steam extraction flow rate. Indicates the reboiler temperature. Indicates regeneration pressure, Indicates flue gas flow rate, To maximize, Indicates minimization; When the variable operating condition is fluctuating flue gas flow rate, the expression for the constraint condition is: ； in, This represents the carbon dioxide capture rate. Indicates flue gas flow rate, The nominal value representing the set flue gas flow rate. Indicates the steam extraction flow rate. This indicates the maximum set extraction steam flow rate. Indicates the reboiler temperature. Indicates regeneration pressure; When the variable operating condition is that the steam extraction capacity is limited, the expression for the constraint condition is: ； in, This represents the carbon dioxide capture rate. Indicates the steam extraction flow rate. This indicates the maximum set extraction steam flow rate. Indicates the reboiler temperature. Indicates regeneration pressure; When the variable operating condition is a capture rate limitation, the expression for the constraint is: ； in, This represents the carbon dioxide capture rate. Indicates the steam extraction flow rate. This indicates the maximum set extraction steam flow rate. Indicates the reboiler temperature. Indicates regeneration pressure; When the variable operating condition is a production constraint, the expression for the constraint is: ； in, This represents the carbon dioxide capture rate. Indicates the steam extraction flow rate. This indicates the maximum set extraction steam flow rate. Indicates the reboiler temperature. Indicates regeneration pressure, Indicates carbon dioxide capture yield. This represents the set minimum carbon dioxide capture yield. This indicates the maximum set carbon dioxide capture yield; The carbon dioxide capture yield is calculated using the following formula: ； in, This represents the concentration of carbon dioxide in the flue gas. This indicates the flue gas flow rate.

2. The performance optimization method for a chemical absorption carbon capture system under varying operating conditions according to claim 1, characterized in that, After obtaining historical operating data of the chemical carbon capture system, key operating parameters affecting the performance of the chemical carbon capture system are screened out using the following formula: ； in, This represents a key data matrix composed of key operating parameters. This represents the original data matrix composed of runtime parameters, with dimensions of [missing information]. , Indicates the number of samples. Indicates the number of runtime parameters. This represents the defined mean vector. This represents the set principal component loading matrix, which is obtained by eigenvalue decomposition of the data covariance matrix of the historical operating data. This indicates transpose.

3. The performance optimization method for a chemical absorption carbon capture system under varying operating conditions according to claim 1, characterized in that, The method of using TOPSIS to perform weighted evaluation of the Pareto front solution set to obtain the optimal solution includes: If the Pareto front solution set contains Given several solutions, construct the decision matrix for each key operating parameter. The expression for the decision matrix is ​​as follows: ； in, Represents the decision matrix. Indicates the first The carbon dioxide capture rate of each solution. Indicates the first The total energy consumption of the chemical absorption carbon capture process for each solution. Indicates the first The total operating cost of a chemical absorption carbon capture system for each solution; The elements in the decision matrix are normalized using the following formula: ； =1,2…, ； =1,2,3； in, express The normalized value, For the first The solution of the first... One goal, when When = 1 corresponds to the carbon dioxide capture rate, when When =2 corresponds to the total energy consumption of the chemical carbon capture process, when The total operating cost of the chemical absorption carbon capture system when the value is 3; The normalized values ​​of each element in the decision matrix are weighted and normalized using the following formula: ； in, express The weighted normalized value, For the first The weight of each objective; The positive and negative ideal solutions are calculated based on the weighted normalized values ​​using the following formula: ； ； in, This represents the positive ideal solution for the carbon dioxide capture rate. express The weighted normalized value, Indicates the first The carbon dioxide capture rate of each solution. The positive ideal solution representing the total energy consumption of the chemical carbon capture process is... Indicates the first The total energy consumption of the chemical absorption carbon capture process for each solution. The positive ideal solution represents the total operating cost of a chemical absorption carbon capture system. Indicates the first The total operating cost of a chemical absorption carbon capture system for a given solution. The negative ideal solution represents the carbon dioxide capture rate. The negative ideal solution represents the total energy consumption of the chemical carbon capture process. The negative ideal solution represents the total operating cost of a chemical absorption carbon capture system. This indicates taking the maximum value. This indicates taking the minimum value; The distance from each solution in the Pareto front solution set to the positive ideal solution is calculated using the following formula: ； in, Indicates the first The distance from each solution to the ideal solution. Indicates the first The positive ideal solution for each objective; The distance from each solution in the Pareto front solution set to the negative ideal solution is calculated using the following formula: ； in, Indicates the first The distance from the ideal solution to the negative ideal solution Indicates the first The negative ideal solution for each objective; The overall score is calculated based on the distances from each solution in the Pareto front solution set to the positive and negative ideal solutions, using the following formula: ； in, Indicates the first The overall score of each solution; choose The largest solution is the optimal solution.

4. A performance optimization device for a chemical absorption carbon capture system under varying operating conditions, characterized in that, include: The key operating parameter screening module is configured to: after acquiring historical operating data of the chemical carbon capture system, screen out the key operating parameters that affect the performance of the chemical carbon capture system; The multi-objective optimization problem construction module is configured to: obtain a chemical carbon capture model based on the mechanism model and physical information neural network of the chemical carbon capture system, and construct a multi-objective optimization problem based on the chemical carbon capture model; The multi-objective optimization problem solving module is configured to: acquire variable operating conditions, determine the constraints of the multi-objective optimization problem based on the variable operating conditions, solve the multi-objective optimization problem based on historical operating data and constraints, and obtain a Pareto front solution set containing the optimal values ​​of each key operating parameter; The optimal solution selection module is configured to use the TOPSIS method to perform a weighted evaluation of the Pareto front solution set to obtain the optimal solution. The performance optimization module is configured to adjust the key operating parameters of the chemical absorption carbon capture system to the values ​​in the optimal solution, thereby completing the performance optimization. The chemical absorption carbon capture model was constructed using the following method: Using the mechanism model of the chemical absorption carbon capture system as the core framework, a physical information neural network is introduced to correct the mechanism model, resulting in a chemical absorption carbon capture model. The aforementioned mechanism model is a mathematical model based on the principles of mass conservation, energy conservation, and chemical reaction equilibrium, describing the chemical absorption carbon capture system under stable operating conditions. The method of introducing a physical information neural network to correct the mechanism model includes: predicting the reaction rate constant affected by operating parameters through the physical information neural network, feeding back the predicted value of the reaction rate constant to the mechanism model, and realizing parameter updates and state correction of the mechanism model under varying operating conditions. The expression for the multi-objective optimization problem is: ； in, This represents the carbon dioxide capture rate. It is the total energy consumption of the chemical carbon capture process. This is the total operating cost of the chemical carbon capture system. Indicates the steam extraction flow rate. Indicates the reboiler temperature. Indicates regeneration pressure, Indicates flue gas flow rate, To maximize, Indicates minimization; When the variable operating condition is fluctuating flue gas flow rate, the expression for the constraint condition is: ； in, This represents the carbon dioxide capture rate. Indicates flue gas flow rate, The nominal value representing the set flue gas flow rate. Indicates the steam extraction flow rate. This indicates the maximum set extraction steam flow rate. Indicates the reboiler temperature. Indicates regeneration pressure; When the variable operating condition is that the steam extraction capacity is limited, the expression for the constraint condition is: ； in, This represents the carbon dioxide capture rate. Indicates the steam extraction flow rate. This indicates the maximum set extraction steam flow rate. Indicates the reboiler temperature. Indicates regeneration pressure; When the variable operating condition is a capture rate limitation, the expression for the constraint is: ； in, This represents the carbon dioxide capture rate. Indicates the steam extraction flow rate. This indicates the maximum set extraction steam flow rate. Indicates the reboiler temperature. Indicates regeneration pressure; When the variable operating condition is a production constraint, the expression for the constraint is: ； in, This represents the carbon dioxide capture rate. Indicates the steam extraction flow rate. This indicates the maximum set extraction steam flow rate. Indicates the reboiler temperature. Indicates regeneration pressure, Indicates carbon dioxide capture yield. This represents the set minimum carbon dioxide capture yield. This indicates the maximum set carbon dioxide capture yield; The carbon dioxide capture yield is calculated using the following formula: ； in, This represents the concentration of carbon dioxide in the flue gas. This indicates the flue gas flow rate.

Images