Method for predicting equilibrium solubility of carbon dioxide absorbed by amine solution

Abstract

The invention discloses a method for predicting the equilibrium solubility of carbon dioxide absorbed by an amine solution. The method comprises the following steps: acquiring a first amine molecule descriptor and a first molecular structure image of a to-be-predicted amine solution; acquiring a second amine molecule descriptor, a second molecular structure image and historical equilibrium solubility of the historical amine solution; constructing an initial graph convolutional neural network model, and inputting the second amine molecule descriptor, the second molecular structure image and the historical equilibrium solubility into the initial graph convolutional neural network model for training to obtain a graph convolutional neural network prediction model; based on the first amine molecule descriptor and the first molecular structure image, prediction is performed through the graph convolutional neural network prediction model, an equilibrium solubility prediction result of the to-be-predicted amine solution is obtained, structure, image and semantic information are fused, and efficient prediction of equilibrium solubility is realized.

Description

Technical Field

[0001] This application relates to the technical field of predicting the equilibrium solubility of amine solutions absorbing carbon dioxide, and in particular to a method for predicting the equilibrium solubility of amine solutions absorbing carbon dioxide. Background Technology

[0002] Global climate change is caused by ever-increasing carbon dioxide emissions. Therefore, many commercial technologies, such as carbon capture, utilization, and storage (CCVS), are crucial for mitigating CO2 emissions. Among these, post-combustion amine chemical absorption has become a mainstream method for large-scale CO2 capture due to its adaptability and technological maturity. However, traditional commercial amine solutions, such as monoethanolamine and diethanolamine, have insufficient absorption capacity under low CO2 partial pressures and high energy consumption during regeneration, limiting their application efficiency. Therefore, the development of alternative amine solutions has attracted attention in industrial applications.

[0003] Currently, there are dozens to hundreds of common amines in laboratories and industry, and the potential types of amine solutions may number in the tens of thousands, not to mention the theoretically countless types of mixed amines. This immense diversity of candidate amines makes manual experimental screening virtually impossible. Summary of the Invention

[0004] This application aims to at least address the technical problems existing in the prior art. To this end, this application proposes a method for predicting the equilibrium solubility of amine solutions absorbing carbon dioxide, which can integrate structural, image, and semantic information to achieve efficient prediction of equilibrium solubility.

[0005] The first aspect of this application provides a method for predicting the equilibrium solubility of an amine solution absorbing carbon dioxide, comprising the following steps:

[0006] Obtain the first amine molecule descriptor and the first molecular structure image of the amine solution to be predicted; obtain the second amine molecule descriptor, the second molecular structure image, and the historical equilibrium solubility of historical amine solutions;

[0007] An initial graph convolutional neural network model is constructed. The second amine molecule descriptor, the second molecule structure image, and the historical equilibrium solubility are input into the initial graph convolutional neural network model for training, thereby obtaining a graph convolutional neural network prediction model.

[0008] Based on the first amine molecule descriptor and the first molecule structure image, the equilibrium solubility prediction result of the amine solution to be predicted is obtained by using the graph convolutional neural network prediction model.

[0009] The method for predicting the equilibrium solubility of amine solutions absorbing carbon dioxide according to the embodiments of this application has at least the following beneficial effects:

[0010] This method acquires the first amine molecule descriptor and the first molecular structure image of the amine solution to be predicted; acquires the second amine molecule descriptor, the second molecular structure image, and the historical equilibrium solubility of historical amine solutions; constructs an initial graph convolutional neural network model, and trains the initial graph convolutional neural network model by inputting the second amine molecule descriptor, the second molecular structure image, and the historical equilibrium solubility to obtain a graph convolutional neural network prediction model; based on the first amine molecule descriptor and the first molecular structure image, the graph convolutional neural network prediction model is used to predict the equilibrium solubility of the amine solution to be predicted, thus integrating structural, image, and semantic information to achieve efficient prediction of equilibrium solubility.

[0011] According to some embodiments of this application, the step of inputting the second amine molecule descriptor, the second molecule structure image, and the historical equilibrium solubility into the initial graph convolutional neural network model for training to obtain a graph convolutional neural network prediction model includes:

[0012] Based on the second amine molecule descriptor, feature extraction is performed through the initial graph convolutional neural network model to obtain the first feature vector;

[0013] Based on the second molecular structure image, feature extraction is performed using the initial graph convolutional neural network model to obtain the spatial relationship feature vector of atomic nodes and bond edges;

[0014] The first feature and the spatial relationship feature are fused to obtain a fused feature vector.

[0015] Based on the fused feature vectors, the initial graph convolutional neural network model is trained to obtain the graph convolutional neural network prediction model.

[0016] According to some embodiments of this application, the step of predicting the equilibrium solubility prediction result of the amine solution to be predicted based on the first amine molecule descriptor and the first molecule structure image using the graph convolutional neural network prediction model includes:

[0017] The first amine molecule descriptor, the first molecule structure image, and the preset absorption conditions are input into the graph convolutional neural network prediction model for prediction to obtain the equilibrium solubility prediction result of the amine solution to be predicted. The preset absorption conditions are at least one of preset absorption temperature, preset carbon dioxide partial pressure, and preset amine solution concentration.

[0018] According to some embodiments of this application, the step of inputting the first amine molecule descriptor, the first molecule structure image, and preset absorption conditions into the graph convolutional neural network prediction model for prediction to obtain the equilibrium solubility prediction result of the amine solution to be predicted includes:

[0019] The first amine molecule descriptor, the first molecule structure image, and the preset absorption conditions are input into the graph convolutional neural network prediction model so that the graph convolutional neural network prediction model calculates the amine molecule similarity using a Gaussian kernel function.

[0020] Based on the amine molecule similarity, the equilibrium solubility prediction result of the amine solution to be predicted is obtained.

[0021] According to some embodiments of this application, obtaining the first amine molecule descriptor of the amine solution to be predicted includes:

[0022] Obtain the SMILES molecular structure code of the amine solution to be predicted;

[0023] The molecular structure SMILES encoding of the molecular structure is analyzed to obtain the analysis results;

[0024] Based on the analysis results, the first amine molecule descriptor of the amine solution to be predicted is extracted, wherein the first amine molecule descriptor includes at least amine molecule topological structure parameters, physicochemical properties and geometric properties.

[0025] According to some embodiments of this application, obtaining the historical equilibrium solubility of a historical amine solution includes:

[0026] Obtain the corresponding equilibrium solubility of historical amine solutions absorbing carbon dioxide under different absorption conditions, wherein the absorption conditions are at least one of absorption temperature, carbon dioxide partial pressure and amine solution concentration;

[0027] The corresponding equilibrium solubility is taken as the historical equilibrium solubility.

[0028] According to some embodiments of this application, the method for predicting the equilibrium solubility of an amine solution absorbing carbon dioxide further includes:

[0029] The amine solution to be predicted is screened based on the preset screening rules and the equilibrium solubility prediction results to obtain the screened amine solution;

[0030] The amine solution after screening was used as the amine solution for the experiment.

[0031] A second aspect of this application provides a system for predicting the equilibrium solubility of an amine solution absorbing carbon dioxide, the system comprising:

[0032] The data acquisition module is used to acquire the first amine molecule descriptor and the first molecular structure image of the amine solution to be predicted; and to acquire the second amine molecule descriptor, the second molecular structure image, and the historical equilibrium solubility of historical amine solutions.

[0033] The training module is used to construct an initial graph convolutional neural network model. The second amine molecule descriptor, the second molecule structure image, and the historical equilibrium solubility are input into the initial graph convolutional neural network model for training, thereby obtaining a graph convolutional neural network prediction model.

[0034] The prediction module is used to make predictions based on the first amine molecule descriptor and the first molecule structure image, using the graph convolutional neural network prediction model, to obtain the equilibrium solubility prediction result of the amine solution to be predicted.

[0035] This system acquires the first amine molecule descriptor and first molecular structure image of the amine solution to be predicted; it also acquires the second amine molecule descriptor, second molecular structure image, and historical equilibrium solubility of historical amine solutions; it constructs an initial graph convolutional neural network model, and trains the initial graph convolutional neural network model by inputting the second amine molecule descriptor, second molecular structure image, and historical equilibrium solubility, thus obtaining a graph convolutional neural network prediction model; based on the first amine molecule descriptor and first molecular structure image, it performs prediction through the graph convolutional neural network prediction model to obtain the equilibrium solubility prediction result of the amine solution to be predicted. By integrating structural, image, and semantic information, it achieves efficient prediction of equilibrium solubility.

[0036] A third aspect of this application provides an electronic device for predicting the equilibrium solubility of an amine solution absorbing carbon dioxide, comprising at least one control processor and a memory for communicatively connecting to the at least one control processor; the memory stores instructions executable by the at least one control processor, the instructions being executed by the at least one control processor to enable the at least one control processor to perform the above-described method for predicting the equilibrium solubility of an amine solution absorbing carbon dioxide.

[0037] A fourth aspect of this application provides a computer-readable storage medium storing computer-executable instructions for causing a computer to execute the above-described method for predicting the equilibrium solubility of an amine solution absorbing carbon dioxide.

[0038] It should be noted that the beneficial effects of the second to fourth aspects of this application compared with the prior art are the same as the beneficial effects of the above-mentioned equilibrium solubility prediction system for amine solution absorbing carbon dioxide compared with the prior art, and will not be described in detail here.

[0039] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0040] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0041] Figure 1 This is a flowchart of a method for predicting the equilibrium solubility of an amine solution absorbing carbon dioxide according to an embodiment of this application;

[0042] Figure 2 This is a schematic diagram of an embodiment of the equilibrium solubility prediction system for amine solution absorbing carbon dioxide provided in this application;

[0043] Figure 3 This is a schematic diagram of the structure of an embodiment of the electronic device provided in this application. Detailed Implementation

[0044] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0045] In the description of this application, the use of terms such as "first," "second," etc., is for the purpose of distinguishing technical features only and should not be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated or the order of the technical features indicated.

[0046] In the description of this application, it should be understood that the orientation descriptions, such as up, down, etc., are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0047] In the description of this application, it should be noted that, unless otherwise explicitly defined, terms such as "setup," "installation," and "connection" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this application in conjunction with the specific content of the technical solution.

[0048] Global climate change is caused by ever-increasing carbon dioxide emissions. Therefore, many commercial technologies, such as carbon capture, utilization, and storage (CCVS), are crucial for mitigating CO2 emissions. Among these, post-combustion amine chemical absorption has become a mainstream method for large-scale CO2 capture due to its adaptability and technological maturity. However, traditional commercial amine solutions, such as monoethanolamine and diethanolamine, have insufficient absorption capacity under low CO2 partial pressures and high energy consumption during regeneration, limiting their application efficiency. Therefore, the development of alternative amine solutions has attracted attention in industrial applications.

[0049] Currently, there are dozens to hundreds of common amines in laboratories and industry, and the potential types of amine solutions may number in the tens of thousands, not to mention the theoretically countless types of mixed amines. This immense diversity of candidate amines makes manual experimental screening virtually impossible.

[0050] To address the aforementioned technical deficiencies, this application provides a method for predicting the equilibrium solubility of an amine solution absorbing carbon dioxide.

[0051] Please see Figure 1 This is a flowchart illustrating a method for predicting the equilibrium solubility of an amine solution absorbing carbon dioxide, provided in an embodiment of this application. This method is applied to electronic devices, such as servers. Figure 1 As shown, the method for predicting the equilibrium solubility of the amine solution absorbing carbon dioxide includes:

[0052] Step S101: Obtain the first amine molecule descriptor and the first molecular structure image of the amine solution to be predicted; obtain the second amine molecule descriptor, the second molecular structure image, and the historical equilibrium solubility of historical amine solutions;

[0053] Step S102: Construct an initial graph convolutional neural network model. Input the second amine molecule descriptor, the second molecule structure image, and the historical equilibrium solubility into the initial graph convolutional neural network model for training to obtain a graph convolutional neural network prediction model.

[0054] Step S103: Based on the first amine molecule descriptor and the first molecule structure image, a graph convolutional neural network prediction model is used to predict the equilibrium solubility of the amine solution to be predicted.

[0055] This method acquires the first amine molecule descriptor and the first molecular structure image of the amine solution to be predicted; acquires the second amine molecule descriptor, the second molecular structure image, and the historical equilibrium solubility of historical amine solutions; constructs an initial graph convolutional neural network model, and trains the initial graph convolutional neural network model by inputting the second amine molecule descriptor, the second molecular structure image, and the historical equilibrium solubility to obtain a graph convolutional neural network prediction model; based on the first amine molecule descriptor and the first molecular structure image, the graph convolutional neural network prediction model is used to predict the equilibrium solubility of the amine solution to be predicted, thus integrating structural, image, and semantic information to achieve efficient prediction of equilibrium solubility.

[0056] In some embodiments, the second amine molecule descriptor, the second molecule structure image, and the historical equilibrium solubility are input into an initial graph convolutional neural network model for training, resulting in a graph convolutional neural network prediction model, including:

[0057] Step S201: Based on the second amine molecule descriptor, feature extraction is performed through the initial graph convolutional neural network model to obtain the first feature vector;

[0058] Step S202: Based on the second molecular structure image, feature extraction is performed using the initial graph convolutional neural network model to obtain the spatial relationship feature vector of atomic nodes and bond edges;

[0059] Step S203: Perform feature fusion on the first feature and the spatial relationship feature to obtain the fused feature vector;

[0060] Step S204: Based on the fused feature vectors, train the initial graph convolutional neural network model to obtain the graph convolutional neural network prediction model.

[0061] Specifically, in some embodiments, this application further includes:

[0062] The determination coefficient, root mean square error, and mean absolute error of the graph convolutional neural network prediction model are calculated based on the prediction results.

[0063] The determination coefficient, root mean square error, and mean absolute error of the trained machine learning model are calculated based on the prediction results. The formulas for calculating the determination coefficient, root mean square error, and mean absolute error include:

[0064]

[0065] Among them, R 2 The coefficient of determination is given by RMSE (root mean square error), MAE (mean absolute error), and N² (number of samples in the test set). This represents the i2th experimental value. This represents the predicted value corresponding to the i2th experimental value. It represents the average of N2 experimental values.

[0066] This application trains an initial graph convolutional neural network model based on the fused feature vector, integrating structural, image, and semantic information, and achieves efficient prediction of balanced solubility.

[0067] In some embodiments, based on the first amine molecule descriptor and the first molecule structure image, a graph convolutional neural network prediction model is used to predict the equilibrium solubility of the amine solution to be predicted, including:

[0068] Step 301: Input the first amine molecule descriptor, the first molecule structure image and the preset absorption conditions into the graph convolutional neural network prediction model for prediction, and obtain the equilibrium solubility prediction result of the amine solution to be predicted. The preset absorption conditions are at least one of the preset absorption temperature, preset carbon dioxide partial pressure and preset amine solution concentration.

[0069] This application improves the accuracy of equilibrium solubility prediction by using preset absorption conditions.

[0070] In some embodiments, the first amine molecule descriptor, the first molecule structure image, and preset absorption conditions are input into a graph convolutional neural network prediction model to obtain the equilibrium solubility prediction result of the amine solution to be predicted, including:

[0071] Step 401: Input the first amine molecule descriptor, the first molecule structure image and the preset absorption conditions into the graph convolutional neural network prediction model so that the graph convolutional neural network prediction model can calculate the amine molecule similarity through the Gaussian kernel function;

[0072] Step 402: Based on the similarity of amine molecules, a prediction is made to obtain the equilibrium solubility prediction result of the amine solution to be predicted.

[0073] The Gaussian kernel function described above is a nonparametric regression method based on the assumption of a joint normal distribution of functions, which involves the concept of Gaussian similarity. An important assumption in Gaussian process regression is that if two x values ​​are similar (e.g., close to each other), then the corresponding y values ​​are also highly correlated. In other words, the covariance matrix is ​​a function of x, and each element of the covariance matrix can be seen as a similarity measure between the corresponding two x values.

[0074] The performance of a recognition system can be severely degraded due to the mismatch between the training and recognition environments, while the Gaussian kernel function can reduce the performance degradation caused by the environment mismatch.

[0075] Specifically, in some embodiments, 12,874 amine substances are crawled from a preset database, and 8,373 potential amine substances are calculated based on the similarity of amine molecules in the dataset.

[0076] Specifically, in some embodiments, the automated machine learning framework for graph convolutional neural network prediction models includes, but is not limited to, H2O.

[0077] H2O, as described above, is an open-source, distributed, memory-based, and scalable automated machine learning framework. It provides a suite of machine learning algorithms, including deep learning, Gradient Boosting Machine (GBM), XGBoost, Random Forest, and Generalized Linear Models (GLM). H2O supports parallel distributed network training, accelerating the model training process, especially when dealing with large amounts of data.

[0078] The equilibrium solubility of all amine solutions that meet the application domain was predicted using a graph convolutional neural network prediction model under fixed absorption conditions (absorption temperature 313 K, CO2 partial pressure 101 kPa, concentration 1.5 mol / L). The prediction results were sorted, screened, and the absorption performance of high-efficiency amine solutions was verified by experiments.

[0079] In some embodiments, obtaining a first amine molecule descriptor for the amine solution to be predicted includes:

[0080] Step 501: Obtain the SMILES encoding of the molecular structure of the amine solution to be predicted;

[0081] Step 502: Perform molecular structure analysis on the SMILES encoding of the molecular structure to obtain the analysis results;

[0082] Step 503: Based on the analysis results, extract the first amine molecule descriptor of the amine solution to be predicted, wherein the first amine molecule descriptor includes at least the amine molecule topological structure parameters, physicochemical properties and geometric properties.

[0083] The SMILES encoding described above is a common pattern for describing molecules using text.

[0084] SMILES strings describe the atoms and bonds of a molecule in a concise and intuitive way for chemists. Each SMILES code corresponds to a unique chemical structure, but each chemical structure can have multiple SMILES code representations. Another important feature of SMILES is its storage efficiency compared to most other methods of representing structures. SMILES takes up 50% to 70% less space than binary tables.

[0085] Furthermore, SMILES compression is highly efficient. With Ziv-Lempel compression, the memory required to store the same database can be reduced to 27% of its original size.

[0086] Specifically, in some embodiments, a total of 2484 datasets were obtained, including amine molecule descriptors, molecular structure images, and equilibrium solubility data of amine solutions absorbing carbon dioxide under different absorption conditions.

[0087] Specifically, in some embodiments, the dataset includes SMILES codes of molecular structures of different primary, secondary, and tertiary amines. Molecular structures are analyzed based on the SMILES codes of amine molecules, and 209 RDKit molecular descriptors, such as amine molecule topological parameters, physicochemical properties, and geometric properties, are extracted based on the analysis results. Different absorption conditions include absorption temperature, carbon dioxide partial pressure, and amine solution concentration.

[0088] The datasets were randomly divided into training and test sets in a ratio of 8 to 2.

[0089] In some embodiments, obtaining the historical equilibrium solubility of a historical amine solution includes:

[0090] Step 601: Obtain the corresponding equilibrium solubility of carbon dioxide absorbed by the amine solution under different absorption conditions, wherein the absorption conditions are at least one of absorption temperature, carbon dioxide partial pressure and amine solution concentration;

[0091] Step 602: Use the corresponding equilibrium solubility as the historical equilibrium solubility.

[0092] This application trains the model by measuring the corresponding equilibrium solubility of carbon dioxide absorbed by historical amine solutions under different absorption conditions, thereby increasing the number of training sets and improving the accuracy of model predictions.

[0093] In some embodiments, the method for predicting the equilibrium solubility of an amine solution absorbing carbon dioxide further includes:

[0094] Step 701: Based on the preset screening rules and the equilibrium solubility prediction results, the amine solution to be predicted is screened to obtain the screened amine solution;

[0095] Step 702: Use the amine solution after screening as the amine solution for the experiment.

[0096] Specifically, for the convenience of those skilled in the art, a set of preferred embodiments is provided below:

[0097] I. Data Acquisition:

[0098] Obtain the first amine molecule descriptor and the first molecular structure image of the amine solution to be predicted; obtain the second amine molecule descriptor, the second molecular structure image, and the historical equilibrium solubility of historical amine solutions, specifically:

[0099] Obtain the SMILES molecular structure encoding of the amine solution to be predicted;

[0100] Molecular structure analysis was performed on the SMILES encoding of the molecular structure to obtain the analysis results;

[0101] Based on the analysis results, the first amine molecule descriptor of the amine solution to be predicted is extracted. The first amine molecule descriptor includes at least the amine molecule topological structure parameters, physicochemical properties and geometric properties.

[0102] Obtain the corresponding equilibrium solubility of carbon dioxide absorbed by the amine solution under different absorption conditions, wherein the absorption conditions are at least one of absorption temperature, carbon dioxide partial pressure and amine solution concentration;

[0103] The corresponding equilibrium solubility is used as the historical equilibrium solubility.

[0104] II. Model Training:

[0105] An initial graph convolutional neural network model is constructed. The descriptor of the second amine molecule, the image of the second molecule structure, and the historical equilibrium solubility are input into the initial graph convolutional neural network model for training, resulting in a graph convolutional neural network prediction model, specifically:

[0106] Based on the second amine molecule descriptor, feature extraction is performed using an initial graph convolutional neural network model to obtain the first feature vector;

[0107] Based on the second molecular structure image, feature extraction is performed using an initial graph convolutional neural network model to obtain spatial relationship feature vectors of atomic nodes and bond edges;

[0108] The first feature and the spatial relationship feature are fused to obtain the fused feature vector.

[0109] Based on the fused feature vectors, the initial graph convolutional neural network model is trained to obtain the graph convolutional neural network prediction model.

[0110] III. Equilibrium Solubility Prediction:

[0111] Based on the first amine molecule descriptor and the first molecule structure image, a graph convolutional neural network prediction model is used to predict the equilibrium solubility of the amine solution to be predicted, specifically:

[0112] The first amine molecule descriptor, the first molecule structure image, and preset absorption conditions are input into a graph convolutional neural network prediction model for prediction, resulting in a predicted equilibrium solubility of the amine solution. The preset absorption conditions are at least one of a preset absorption temperature, a preset carbon dioxide partial pressure, and a preset amine solution concentration, specifically:

[0113] The first amine molecule descriptor, the first molecule structure image, and the preset absorption conditions are input into the graph convolutional neural network prediction model so that the graph convolutional neural network prediction model can calculate the amine molecule similarity through the Gaussian kernel function.

[0114] Based on the similarity of amine molecules, the equilibrium solubility prediction results of the amine solution to be predicted are obtained.

[0115] IV. Screening with amine solutions in experiments:

[0116] Based on the preset screening rules and the equilibrium solubility prediction results, the amine solutions to be predicted are screened to obtain the screened amine solutions;

[0117] The amine solution after screening was used as the amine solution for the experiment.

[0118] Additionally, refer to Figure 2One embodiment of this application provides a system for predicting the equilibrium solubility of an amine solution absorbing carbon dioxide, comprising a data acquisition module 1100, a training module 1200, and a prediction module 1300, wherein:

[0119] The data acquisition module 1100 is used to acquire the first amine molecule descriptor and the first molecular structure image of the amine solution to be predicted; and to acquire the second amine molecule descriptor, the second molecular structure image, and the historical equilibrium solubility of historical amine solutions.

[0120] The training module 1200 is used to construct an initial graph convolutional neural network model. The second amine molecule descriptor, the second molecule structure image, and the historical equilibrium solubility are input into the initial graph convolutional neural network model for training, and a graph convolutional neural network prediction model is obtained.

[0121] The prediction module 1300 is used to make predictions based on the first amine molecule descriptor and the first molecule structure image, using a graph convolutional neural network prediction model, to obtain the equilibrium solubility prediction result of the amine solution to be predicted.

[0122] This system acquires the first amine molecule descriptor and first molecular structure image of the amine solution to be predicted; it also acquires the second amine molecule descriptor, second molecular structure image, and historical equilibrium solubility of historical amine solutions; it constructs an initial graph convolutional neural network model, and trains the initial graph convolutional neural network model by inputting the second amine molecule descriptor, second molecular structure image, and historical equilibrium solubility, thus obtaining a graph convolutional neural network prediction model; based on the first amine molecule descriptor and first molecular structure image, it performs prediction through the graph convolutional neural network prediction model to obtain the equilibrium solubility prediction result of the amine solution to be predicted. By integrating structural, image, and semantic information, it achieves efficient prediction of equilibrium solubility.

[0123] It should be noted that the system embodiments described above are based on the same inventive concept as the method embodiments described above. Therefore, the relevant content of the method embodiments described above is also applicable to the system embodiments described above, and will not be repeated here.

[0124] Figure 3 A schematic diagram of the rule mining hardware structure provided in an embodiment of this application is shown.

[0125] The equilibrium solubility prediction device for amine solution absorbing carbon dioxide may include a processor 301 and a memory 302 storing computer program instructions.

[0126] Specifically, the processor 301 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0127] Memory 302 may include mass storage for data or instructions. For example, and not limitingly, memory 302 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 302 may include removable or non-removable (or fixed) media. Where appropriate, memory 302 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 302 is non-volatile solid-state memory.

[0128] In some embodiments, memory 302 may include read-only memory (ROM), random access memory (RAM), disk storage media device, optical storage media device, flash memory device, electrical, optical, or other physical / tangible memory storage device. Thus, generally, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to one aspect of this disclosure.

[0129] The processor 301 reads and executes computer program instructions stored in the memory 302 to implement any of the equilibrium solubility prediction methods for amine solutions absorbing carbon dioxide in the above embodiments.

[0130] In one example, the equilibrium solubility prediction device for amine solution absorbing carbon dioxide may further include a communication interface 303 and a bus 310. Wherein, as Figure 3 As shown, the processor 301, memory 302, and communication interface 303 are connected through bus 310 and complete communication with each other.

[0131] The communication interface 303 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0132] Bus 310 includes hardware, software, or both, that couples together components of an equilibrium solubility prediction device for amine solution absorbing carbon dioxide. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 310 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, any suitable bus or interconnect is contemplated herein.

[0133] The equilibrium solubility prediction device for amine solution absorbing carbon dioxide can execute the equilibrium solubility prediction method for amine solution absorbing carbon dioxide in the embodiments of this application based on a three-dimensional design model, thereby achieving a combination of... Figure 1 and Figure 2 A method and system for predicting the equilibrium solubility of amine solutions absorbing carbon dioxide are described.

[0134] Furthermore, in conjunction with the equilibrium solubility prediction method for amine solution absorbing carbon dioxide in the above embodiments, this application embodiment can provide a computer storage medium for implementation. This computer storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the equilibrium solubility prediction methods for amine solution absorbing carbon dioxide in the above embodiments.

[0135] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0136] The functional blocks shown in the above block diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0137] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0138] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block in 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, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0139] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A method for predicting the equilibrium solubility of an amine solution absorbing carbon dioxide, characterized in that, The method for predicting the equilibrium solubility of the amine solution in the absorption of carbon dioxide includes: Obtain the first amine molecule descriptor and the first molecular structure image of the amine solution to be predicted; obtain the second amine molecule descriptor, the second molecular structure image, and the historical equilibrium solubility of historical amine solutions; An initial graph convolutional neural network model is constructed. The second amine molecule descriptor, the second molecule structure image, and the historical equilibrium solubility are input into the initial graph convolutional neural network model for training, thereby obtaining a graph convolutional neural network prediction model. Based on the first amine molecule descriptor and the first molecule structure image, the equilibrium solubility prediction result of the amine solution to be predicted is obtained by using the graph convolutional neural network prediction model.

2. The method for predicting the equilibrium solubility of amine solution absorbing carbon dioxide according to claim 1, characterized in that, The step of inputting the second amine molecule descriptor, the second molecule structure image, and the historical equilibrium solubility into the initial graph convolutional neural network model for training to obtain a graph convolutional neural network prediction model includes: Based on the second amine molecule descriptor, feature extraction is performed through the initial graph convolutional neural network model to obtain the first feature vector; Based on the second molecular structure image, feature extraction is performed using the initial graph convolutional neural network model to obtain the spatial relationship feature vector of atomic nodes and bond edges; The first feature and the spatial relationship feature are fused to obtain a fused feature vector. Based on the fused feature vectors, the initial graph convolutional neural network model is trained to obtain the graph convolutional neural network prediction model.

3. The method for predicting the equilibrium solubility of amine solution absorbing carbon dioxide according to claim 1, characterized in that, The step of predicting the equilibrium solubility of the amine solution to be predicted by using the graph convolutional neural network prediction model based on the first amine molecule descriptor and the first molecule structure image includes: The first amine molecule descriptor, the first molecule structure image, and the preset absorption conditions are input into the graph convolutional neural network prediction model for prediction to obtain the equilibrium solubility prediction result of the amine solution to be predicted. The preset absorption conditions are at least one of preset absorption temperature, preset carbon dioxide partial pressure, and preset amine solution concentration.

4. The method for predicting the equilibrium solubility of amine solution absorbing carbon dioxide according to claim 3, characterized in that, The step of inputting the first amine molecule descriptor, the first molecule structure image, and preset absorption conditions into the graph convolutional neural network prediction model for prediction to obtain the equilibrium solubility prediction result of the amine solution to be predicted includes: The first amine molecule descriptor, the first molecule structure image, and the preset absorption conditions are input into the graph convolutional neural network prediction model so that the graph convolutional neural network prediction model calculates the amine molecule similarity using a Gaussian kernel function. Based on the amine molecule similarity, the equilibrium solubility prediction result of the amine solution to be predicted is obtained.

5. The method for predicting the equilibrium solubility of amine solution absorbing carbon dioxide according to claim 1, characterized in that, The process of obtaining the first amine molecule descriptor for the amine solution to be predicted includes: Obtain the SMILES molecular structure code of the amine solution to be predicted; The molecular structure SMILES encoding of the molecular structure is analyzed to obtain the analysis results; Based on the analysis results, the first amine molecule descriptor of the amine solution to be predicted is extracted, wherein the first amine molecule descriptor includes at least amine molecule topological structure parameters, physicochemical properties and geometric properties.

6. The method for predicting the equilibrium solubility of amine solution absorbing carbon dioxide according to claim 1, characterized in that, Obtain the historical equilibrium solubility of the amine solution, including: Obtain the corresponding equilibrium solubility of historical amine solutions absorbing carbon dioxide under different absorption conditions, wherein the absorption conditions are at least one of absorption temperature, carbon dioxide partial pressure and amine solution concentration; The corresponding equilibrium solubility is taken as the historical equilibrium solubility.

7. The method for predicting the equilibrium solubility of amine solution absorbing carbon dioxide according to claim 1, characterized in that, The method for predicting the equilibrium solubility of amine solutions absorbing carbon dioxide also includes: The amine solution to be predicted is screened based on the preset screening rules and the equilibrium solubility prediction results to obtain the screened amine solution; The amine solution after screening was used as the amine solution for the experiment.

8. A system for predicting the equilibrium solubility of an amine solution absorbing carbon dioxide, characterized in that, The equilibrium solubility prediction system for carbon dioxide absorption by the amine solution includes: The data acquisition module is used to acquire the first amine molecule descriptor and the first molecular structure image of the amine solution to be predicted; and to acquire the second amine molecule descriptor, the second molecular structure image, and the historical equilibrium solubility of historical amine solutions. The training module is used to construct an initial graph convolutional neural network model. The second amine molecule descriptor, the second molecule structure image, and the historical equilibrium solubility are input into the initial graph convolutional neural network model for training, thereby obtaining a graph convolutional neural network prediction model. The prediction module is used to make predictions based on the first amine molecule descriptor and the first molecule structure image, using the graph convolutional neural network prediction model, to obtain the equilibrium solubility prediction result of the amine solution to be predicted.

9. A device for predicting the equilibrium solubility of an amine solution absorbing carbon dioxide, characterized in that, It includes at least one control processor and a memory for communicatively connecting to the at least one control processor; the memory stores instructions executable by the at least one control processor to enable the at least one control processor to perform a method for predicting the equilibrium solubility of an amine solution absorbing carbon dioxide as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions for causing a computer to perform a method for predicting the equilibrium solubility of an amine solution absorbing carbon dioxide as described in any one of claims 1 to 7.

Images