Rapid prediction method and system for thermal weight loss property of chemical-looping oxygen carrier material based on data driving, medium and program product

Abstract

The invention relates to a data-driven rapid prediction method and system for the thermal weight loss property of a chemical-looping oxygen carrier material, a medium and a program product, and the method comprises the steps: building a machine learning model, taking the structural characteristics and temperature rise rate of the chemical-looping oxygen carrier material as input, achieving the prediction of a material differential quotient thermogravimetric curve on one hand, and achieving the rapid prediction of the thermal weight loss property of the chemical-looping oxygen carrier material on the other hand; a thermogravimetric curve of the material is obtained based on the differential quotient thermogravimetric curve of the material, and then the thermal weight loss property is obtained through analysis; on the other hand, prediction of material thermal weight loss properties such as the oxygen decoupling temperature and the oxygen decoupling mass fraction of the chemical chain oxygen carrier is directly achieved. Compared with the prior art, the screening efficiency of the chemical chain oxygen carrier is greatly improved, the screening method has the technical advantages of being large in batch, low in cost, high in precision and the like, and the problems that traditional material experiment testing is time-consuming and labor-consuming and the like are solved; and meanwhile, the method can be expanded and applied to metal oxide oxygen decoupling temperature identification and performance prediction in other application scenes such as thermochemical energy storage and the like.

Description

Technical Field

[0001] The present invention relates to the field of chemical looping combustion, and in particular, to a method, system, medium, and program product for rapidly predicting the thermogravimetric properties of chemical looping oxygen carriers based on data-driven Background Art

[0002] Chemical looping combustion (CLC) technology is a new combustion method for in-situ capturing CO 2 The principle is that metal oxides are used as oxygen carriers to absorb O from the air 2 And transfer it to the fuel, so as to realize the in-situ separation and enrichment utilization of CO during the combustion process 2 Compared with traditional chemical looping combustion technology, chemical looping with oxygen uncoupling (CLOU) technology requires the oxygen uncoupling performance of oxygen carriers. Metal oxides decouple oxygen molecules at a certain temperature to directly participate in the combustion and conversion reactions of hydrocarbon fuels. Oxygen uncoupling can effectively promote the reaction rate of fuel gasification and combustion, alleviate the problem of carbon deposition on the oxygen carrier volume, and thus improve the overall efficiency of the CLC process. Therefore, the research and development of oxygen uncoupling oxygen carriers is a necessary way to realize CLOU technology. Rapid identification and accurate prediction of the thermogravimetric properties of metal oxides are the key breakthrough points for designing high-performance CLOU oxygen carrier materials

[0003] At present, the research on the thermogravimetric properties of oxygen carrier materials mainly relies on thermogravimetric experiments. By heating at a programmed rate to obtain the weight loss curve, and then obtaining the analysis of the thermogravimetric properties. However, thermogravimetric experiments are time-consuming and costly. Moreover, oxygen carrier materials are mainly composed of multi-metal oxides. With their huge element combinations and structural types, the accurate identification and directional design of high-performance oxygen uncoupling oxygen carriers are still unresolved problems for traditional experimental trial-and-error models Summary of the Invention

[0004] The purpose of the present invention is to overcome the above-mentioned defects existing in the prior art and provide a method, system, medium, and program product for rapidly predicting the thermogravimetric properties of chemical looping oxygen carriers based on data-driven. Compared with traditional thermogravimetric experimental research methods, it has low cost and high reliability, and helps to rapidly and accurately identify the thermogravimetric properties of chemical looping oxygen carrier materials

[0005] The purpose of the present invention can be achieved by the following technical solutions

[0006] The present invention provides a method for rapidly predicting the thermogravimetric properties of chemical looping oxygen carriers based on data-driven, including the following steps

[0007] S1: Using X-ray diffraction data as the characteristic descriptor of the chemical chain oxygen carrier material structure, using the differential thermogravimetric curve and thermal weight loss properties as the characteristic descriptors of the chemical chain oxygen carrier material performance, using the characteristic descriptor of the material structure and the temperature rise rate as input, and using the characteristic descriptor of the material performance as output, a machine learning model is constructed, and through training, a differential thermogravimetric curve prediction model and a direct prediction model of thermal weight loss properties for the chemical chain oxygen carrier material are obtained respectively;

[0008] S2: Based on the characteristics of the target material structure and the temperature rise rate, on the one hand, the material derivative thermogravimetric curve is obtained by using the trained derivative thermogravimetric curve prediction model, and based on the predicted material derivative thermogravimetric curve, the derivative thermogravimetric is converted into thermogravimetric through integration, and the derivative thermogravimetric and thermogravimetric curves are analyzed to achieve the analysis and prediction of the thermal weight loss properties of the chemical chain oxygen carrier material; on the other hand, the trained direct prediction model of thermal weight loss properties is used to directly obtain the prediction of the thermal weight loss properties of the chemical chain oxygen carrier material.

[0009] Further, in S1, the X-ray diffraction data and the derivative thermogravimetric curve include:

[0010] For X-ray diffraction data, the range of diffraction angles was unified to 10-85°, with an angle step of 0.05°, and the range of diffraction intensity data was unified to 0-100, with a total of 1500 sets of data points with one-to-one correspondence between angles and intensities, thus achieving structural unification of X-ray diffraction data;

[0011] For the Weishang thermogravimetric curve, the temperature range was unified to 700-1000 °C, with a total of 30 sets of data points with one-to-one correspondence between temperature and mass change rate.

[0012] Further, in S1, the thermal weight loss properties include:

[0013] Oxygen decoupling temperature, oxygen decoupling mass fraction, wherein the oxygen decoupling temperature includes the initial oxygen decoupling temperature and the optimal oxygen decoupling temperature.

[0014] Furthermore, in S1, the machine learning model construction includes:

[0015] Machine learning models are constructed to realize the prediction of the derivative thermogravimetric curve and the direct prediction of the thermal weight loss properties of chemical chain oxygen carrier materials. The models are used to mine the mapping relationship between the X-ray diffraction data, temperature rise rate and derivative thermogravimetric curve, and thermal weight loss properties of oxygen carrier materials. The machine learning model includes one or more of convolutional neural networks, artificial neural networks, and Transformer.

[0016] Furthermore, the construction of the Transformer model includes:

[0017] The X-ray diffraction data is equally divided into 30 parts according to the diffraction angle, with each part containing 50 groups of data points. The corresponding diffraction intensity information is converted into a 30×50 sequence form, where 30 is the time step and 50 is the feature dimension, serving as the input format for the Transformer model. After the X-ray diffraction data is embedded through a linear layer, positional encoding based on sine and cosine functions is added to add positional information to the sequence data, and two Transformer Encoder layers are used as the encoder for encoding. For the encoded data, the input of the temperature rise rate information is added, and then a single fully connected layer is used as the decoder for decoding. Finally, it is mapped to the data points of the derivative thermogravimetric curve in a multi-output form and to the thermogravimetric property in a single-output form.

[0018] Further, in S2, the analysis and prediction of the thermogravimetric property based on derivative thermogravimetry and thermogravimetry include:

[0019] The data points where the mass change rate of the derivative thermogravimetric curve shows a significant increase are defined as the initial oxygen decoupling points, and the corresponding temperature is the initial oxygen decoupling temperature; the highest point of the derivative thermogravimetric curve is defined as the optimal oxygen decoupling point, and the corresponding temperature is the optimal oxygen decoupling temperature.

[0020] Integral calculation is performed on the derivative thermogravimetric curve to obtain the thermogravimetric curve, and the difference between the mass percentages corresponding to the initial oxygen decoupling temperature and the final temperature is denoted as the oxygen decoupling mass fraction.

[0021] The second aspect of the present invention provides a rapid prediction system for the thermogravimetric property of a chemical-looping oxygen carrier material based on data driving, including:

[0022] A model construction module uses X-ray diffraction data as the feature descriptor of the material structure, derivative thermogravimetric curve and thermogravimetric property as the feature descriptors of the material performance, uses the structural features and temperature rise rate as the input, and the performance features as the output to construct a machine learning model, respectively realizing the prediction of the derivative thermogravimetric curve of the chemical-looping oxygen carrier material and the direct prediction of the thermogravimetric property.

[0023] A performance prediction module, on the one hand, inputs the X-ray diffraction data of the material to obtain the prediction of the derivative thermogravimetric curve of the material, obtains the thermogravimetric curve based on the derivative thermogravimetry, and further analyzes and obtains the oxygen decoupling temperature and oxygen decoupling mass fraction of the material, realizing the rapid analysis and prediction of the thermogravimetric property of the chemical-looping oxygen carrier material; on the other hand, inputs the structural data of the material to directly obtain the prediction of the thermogravimetric property data.

[0024] Further, the rapid prediction method and system for the thermogravimetric property of the chemical-looping oxygen carrier material provided by the present invention can be extended and applied to the identification and performance prediction of application scenarios related to metal oxide oxygen decoupling.

[0025] The oxygen decoupling process of metal oxides can be applied to different low-carbon utilization technology scenarios, such as Chemical looping with oxygen uncoupling (CLOU), Thermochemical Energy Storage (TCES), Thermochemical watersplitting (TCWS), etc. There are differences in the requirements for the thermogravimetric properties of metal oxide materials in different application scenarios. Based on the established prediction model for the thermogravimetric properties of chemical looping oxygen carriers, by inputting the X-ray diffraction data of metal oxides, the oxygen decoupling temperature is output, and the applicable application scenarios of metal oxide materials are identified based on the classification of the starting oxygen decoupling temperature.

[0026] The third aspect of the present invention provides a storage medium containing computer-executable instructions, which is used to execute the rapid prediction method for the thermogravimetric properties of chemical looping oxygen carriers based on data as described above when executed by a computer processor.

[0027] The fourth aspect of the present invention provides a computer program product, including a computer program, which realizes the rapid prediction method for the thermogravimetric properties of chemical looping oxygen carriers based on data as described above when executed by a processor.

[0028] Compared with the prior art, the present invention has the following technical advantages:

[0029] 1) By constructing machine learning to explore the mapping relationship between the structural data of chemical looping oxygen carriers and the derivative thermogravimetric curve and thermogravimetric properties, the identification ability of chemical looping oxygen carriers is significantly improved. Compared with traditional experimental tests, the method of the present invention is more objective and accurate, providing a powerful tool for researchers in the field of materials science.

[0030] 2) The present invention realizes the rapid identification of oxygen carrier materials with oxygen decoupling characteristics, enabling researchers to quickly predict and analyze the thermogravimetric properties of oxygen carrier materials. This method not only saves valuable time and human resources, but also reduces the consumption of materials and energy, conforms to the concept of green chemistry and sustainable development, and provides an important scientific basis and technical support for promoting the progress of energy conversion and storage technologies. Description of the Drawings

[0031] Figure 1 It is a working flow chart of the rapid prediction method for the thermogravimetric properties of chemical looping oxygen carriers based on data provided by an embodiment of the present invention;

[0032] Figure 2Schematic structural diagram of the Transformer model provided according to an embodiment of the present invention;

[0033] Figure 3 Comparison diagram of the predicted value and the true value of the derivative thermogravimetric curve of some materials in the training set provided according to an embodiment of the present invention;

[0034] Figure 4 Comparison diagram of the predicted value and the true value of the derivative thermogravimetric curve of some materials in the test set provided according to an embodiment of the present invention;

[0035] Figure 5 Comparison diagram of the true value and the predicted value of the initial oxygen decoupling temperature of the material obtained based on the derivative thermogravimetric curve analysis provided according to an embodiment of the present invention;

[0036] Figure 6 Comparison diagram of the true value and the predicted value of the optimal oxygen decoupling temperature of the material obtained based on the derivative thermogravimetric curve analysis provided according to an embodiment of the present invention;

[0037] Figure 7 Comparison diagram of the true value and the predicted value of the thermogravimetric curve of some materials obtained based on the derivative thermogravimetric curve analysis provided according to an embodiment of the present invention;

[0038] Figure 8 Comparison diagram of the true value and the predicted value of the oxygen decoupling mass fraction of the material obtained based on the thermogravimetric curve analysis;

[0039] Figure 9 Comparison diagram of the true value and the predicted value of the initial oxygen decoupling temperature of the material provided according to an embodiment of the present invention;

[0040] Figure 10 Comparison diagram of the true value and the predicted value of the oxygen decoupling mass fraction of the material provided according to an embodiment of the present invention. Detailed implementation manners

[0041] Overall, the present invention provides a rapid prediction method, system, and medium for the thermogravimetric properties of chemical-looping oxygen carrier materials based on data-driven. The specific method is as follows: Using X-ray diffraction data as the characteristic descriptor of the oxygen carrier material structure, and the derivative thermogravimetric curve as the characteristic descriptor of the material performance. Taking the structural characteristics and temperature rise rate as inputs and the performance characteristics as outputs, a machine learning model is constructed to achieve the prediction of the derivative thermogravimetric curve. Based on the derivative thermogravimetric curve of the material, the oxygen decoupling temperature and oxygen decoupling mass fraction data of the material are further analyzed to achieve the analysis and prediction of the thermogravimetric properties of the chemical-looping oxygen carrier materials. The present invention constructs a mapping relationship between the structural data of the oxygen carrier material and the derivative thermogravimetric curve through a machine learning model, with low cost and high reliability, realizes the rapid identification of oxygen carrier materials with oxygen decoupling characteristics, and avoids problems such as time-consuming and laborious traditional material experimental tests.

[0042] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. Features such as component models, material names, connection structures, control methods, algorithms, etc. that are not clearly described in this technical solution are regarded as common technical features disclosed in the prior art.

[0043] Example 1

[0044] This example discloses a rapid prediction method for the thermogravimetric properties of chemical-looping oxygen carrier materials based on data-driven, referring to Figure 1 as shown, and specifically includes the following steps:

[0045] S1: Using X-ray diffraction data as the characteristic descriptor of the chemical-looping oxygen carrier material structure, and the derivative thermogravimetric curve and thermogravimetric properties as the characteristic descriptors of the chemical-looping oxygen carrier material performance. Taking the characteristic descriptor of the material structure and the temperature rise rate as inputs and the characteristic descriptor of the material performance as outputs, a machine learning model is constructed, and through training, a prediction model for the derivative thermogravimetric curve of the chemical-looping oxygen carrier material and a direct prediction model for the thermogravimetric properties are obtained respectively.

[0046] S2: Based on the characteristics of the target material structure and the temperature rise rate, on the one hand, using the trained derivative thermogravimetric curve prediction model to obtain the material derivative thermogravimetric curve, and based on the predicted material derivative thermogravimetric curve, converting the derivative thermogravimetric into thermogravimetric by integration, analyzing the derivative thermogravimetric and thermogravimetric curves to realize the analysis and prediction of the thermogravimetric properties of the chemical-looping oxygen carrier material. On the other hand, using the trained direct prediction model of the thermogravimetric properties, directly obtaining the prediction of the thermogravimetric properties of the chemical-looping oxygen carrier material. Specifically, in step S1, the structuring of the X-ray diffraction data and the acquisition of the oxygen decoupling temperature include:

[0047] For X-ray diffraction data, the range of diffraction angles was unified to 10-85°, with an angle step of 0.05°, and the range of diffraction intensity data was unified to 0-100, with a total of 1500 sets of data points with one-to-one correspondence between angles and intensities, thus achieving structural unification of X-ray diffraction data;

[0048] For the Weishang thermogravimetric curve, the temperature range was unified to 700-1000 °C, with a total of 30 sets of data points with one-to-one correspondence between temperature and mass change rate.

[0049] In a specific implementation, in step S1, the machine learning model is constructed, including:

[0050] A machine learning model is constructed to predict the differential thermogravimetric curve of chemical chain oxygen carrier materials. The model is used to mine the mapping relationship between the X-ray diffraction data, temperature rise rate and differential thermogravimetric curve of oxygen carrier materials. The machine learning model includes one or more of convolutional neural networks, artificial neural networks, random forests and Transformers. In particular, the structural diagram of the Transformer model is shown in Figure 2 shown.

[0051] In the specific implementation, in this step, X-ray diffraction data and temperature rise rate are used as model input, and differential thermogravimetric curve data are used as model output to build a differential thermogravimetric curve prediction model based on Transformer. All data are 28 groups in total, divided into training set and test set in a ratio of 8:2. The training set is used to train the model, and the test set is used to verify the model prediction effect. After the model is trained, based on R 2 Or the mean absolute error MAE is used for model evaluation, and the calculation formula is as follows:

[0052]

[0053] Among them, y i is the true value, is the predicted value, is the average value.

[0054] The results show that the prediction effect of the trained Transformer model on the differential thermogravimetric curve is: the average R 2 The average R 2 The comparison diagrams of the differential thermogravimetric curves of some materials predicted by the model in the training set and the test set and the true values ​​are shown in Figure 3 , Figure 4 As shown, the materials listed include: CuO, different MgO, Al 2 O 3Proportion-modified CuO (Cu90MgAl-1000, Cu70MgAl-1000, Cu90Al-1000), 60% cerium-doped CuO (Cu60Ce), and ZrO 2 supported CuO (CuO-ZrO 2 ).

[0055] In specific implementation, based on the analysis of the derivative thermogravimetric curve, the prediction of the oxygen decoupling temperature is realized. The model predicts the corresponding derivative thermogravimetric curve, and further analyzes each derivative thermogravimetric curve to obtain the oxygen decoupling temperature of the corresponding material, including the initial oxygen decoupling temperature and the optimal oxygen decoupling temperature. The comparison between the true value and the predicted value of the initial oxygen decoupling temperature is as Figure 5 shown, and the comparison between the true value and the predicted value of the optimal oxygen decoupling temperature is as Figure 6 shown. The mean absolute error between the true value and the predicted value of the initial oxygen decoupling temperature is 7.73 °C on the training set and 15.00 °C on the test set; the mean absolute error between the true value and the predicted value of the optimal oxygen decoupling temperature is 2.27 °C on the training set and 8.33 °C on the test set.

[0056] In specific implementation, the derivative thermogravimetric curve is transformed into a thermogravimetric curve through an integration operation, and further the prediction of the oxygen decoupling mass fraction is realized. The comparison between the true value and the predicted value of the thermogravimetric curve of some materials is as Figure 7 shown. The model predicts the corresponding derivative thermogravimetric curve, and further integrates each derivative thermogravimetric curve into a thermogravimetric curve, so as to obtain the oxygen decoupling mass fraction data of the corresponding material. The comparison between the true value and the predicted value of the oxygen decoupling mass fraction is as Figure 8 shown. The mean absolute error between the true value and the predicted value of the oxygen decoupling mass fraction is 0.24 on the training set and 0.8 on the test set.

[0057] In this embodiment, by constructing a machine learning model to mine the mapping relationship between the material structure data, the temperature rise rate and the derivative thermogravimetric curve of the chemical-looping oxygen carrier, based on the derivative thermogravimetric curve of the material, the analysis and prediction of the thermal weight loss property of the chemical-looping oxygen carrier material are realized. The model greatly improves the identification ability of the thermal weight loss property of the oxygen carrier material, and avoids the problems such as time-consuming and laborious of traditional material experimental tests.

[0058] This embodiment of the present invention also discloses a rapid prediction system for the thermal weight loss property of a chemical-looping oxygen carrier material based on data driving, specifically including:

[0059] A model construction module, which uses X-ray diffraction data as the feature descriptor of the material structure, uses the derivative thermogravimetric curve as the feature descriptor of the material performance, uses the structural features and the temperature rise rate as the input, and the performance features as the output, constructs a machine learning model, and realizes the direct prediction of the derivative thermogravimetric curve of the chemical-looping oxygen carrier material.

[0060] A performance prediction module, based on the analysis of the derivative thermogravimetric curve of the material, obtains the oxygen decoupling temperature and oxygen decoupling mass fraction data of the material, and realizes the rapid analysis and prediction of the thermal weight loss properties of the chemical looping oxygen carrier material.

[0061] The modules described as separation components above may or may not be physically separated. The components shown as modules may or may not be physical modules, that is, they may be located in one place, or they may be distributed to multiple network modules. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0062] In addition, in each embodiment of the present invention, the functional modules can be integrated in a processing module, or each module can exist physically alone, or two or more modules can be integrated in one module. The above integrated modules can be implemented in the form of hardware or in the form of software functional units.

[0063] This embodiment also proposes a computer-readable storage medium, which stores computer instructions for causing a computer to execute the above-mentioned rapid prediction method for the thermal weight loss properties of the chemical looping oxygen carrier material based on data. The storage medium can be an electronic medium, a magnetic medium, an optical medium, an electromagnetic medium, an infrared medium or a semiconductor system or a propagation medium. The storage medium can also include semiconductor or solid-state memory, magnetic tape, removable computer disk, random access memory (RAM), read-only memory (ROM), hard disk and optical disk. The optical disk can include compact disc-read only memory (CD-ROM), compact disc-rewritable (CD-RW) and DVD.

[0064] This embodiment also provides a computer program product, including a computer program, which when executed by a processor implements the above-mentioned rapid prediction method for the thermal weight loss properties of the chemical looping oxygen carrier material based on data. When this computer program product runs on various suitable processors, it can quickly and accurately call the required computing resources, and process the input data related to the chemical looping oxygen carrier material, such as X-ray diffraction data, temperature rise rate, etc. according to the established process. By executing the constructed machine learning model, whether it is a convolutional neural network, an artificial neural network or a Transformer model, it can realize the mapping calculation from the material structure characteristics to the performance characteristics, and then accurately predict the derivative thermogravimetric curve and thermal weight loss properties of the chemical looping oxygen carrier material. Moreover, this computer program product also has good compatibility and can be adapted to a variety of operating systems and hardware platforms

[0065] Embodiment 2

[0066] This embodiment discloses a method for rapidly predicting the oxygen decoupling temperature of a chemical-looping oxygen carrier material based on data-driven, which specifically includes the following steps:

[0067] Taking X-ray diffraction data as the characteristic descriptor of the material structure, the initial oxygen decoupling temperature as the characteristic descriptor of the material performance, the structural characteristics and the temperature rise rate as the inputs, and the initial oxygen decoupling temperature as the output, a Transformer model for directly predicting the initial oxygen decoupling temperature of the chemical-looping oxygen carrier material is constructed using machine learning.

[0068] There are a total of 80 groups of X-ray diffraction data and initial oxygen decoupling temperature data of the chemical-looping oxygen carrier material, which are divided into a training set and a test set according to a ratio of 9:1. The training set is used to train the model, and the test set is used to verify the model effect; after the model is trained, based on R 2 Model evaluation is carried out.

[0069] The results show that the prediction effect of the trained Transformer model on the initial oxygen decoupling temperature is: R of the training set 2 is 0.767, and R of the test set 2 is 0.636. The comparison diagram of the predicted values and the true values of the initial oxygen decoupling temperature of the material on the training set and the test set is as Figure 9 shown.

[0070] Example 3

[0071] This embodiment discloses a method for rapidly predicting the oxygen decoupling mass fraction of a chemical-looping oxygen carrier material, which specifically includes:

[0072] Taking X-ray diffraction data as the characteristic descriptor of the material structure, the oxygen decoupling mass fraction as the characteristic descriptor of the material performance, the structural characteristics and the temperature rise rate as the inputs, and the oxygen decoupling mass fraction as the output, a Transformer model for directly predicting the oxygen decoupling mass fraction of the chemical-looping oxygen carrier material is constructed using machine learning.

[0073] There are a total of 130 groups of X-ray diffraction data and oxygen decoupling mass fraction data of the chemical-looping oxygen carrier material, which are divided into a training set and a test set according to a ratio of 9:1. The training set is used to train the model, and the test set is used to verify the model effect; after the model is trained, based on R 2 Model evaluation is carried out.

[0074] The results show that the prediction effect of the trained Transformer model on the oxygen decoupling mass fraction is: R of the training set 2 is 0.988, and R of the test set 2 is 0.890. The comparison diagram of the predicted values and the true values of the oxygen decoupling mass fraction of the material on the training set and the test set is asFigure 10 as shown

[0075] Example 4

[0076] This example discloses a classification and identification based on the initial oxygen decoupling temperature, realizing a method for identifying application scenarios related to the oxygen decoupling process of metal oxides, specifically including the following steps:

[0077] Using the derivative thermogravimetric curve prediction model for chemical looping oxygen carrier materials constructed in Example 1, the oxygen decoupling temperature is analyzed, or using the thermogravimetric property prediction model constructed in Example 2, the initial oxygen decoupling temperature of the material is directly obtained, and further the application scenarios of the metal oxide material oxygen decoupling process are identified. There are a total of 164 groups of structural and performance data of different metal oxide materials studied under the CLOU scenario, TCES scenario, and TCWS scenario. By inputting their X-ray diffraction data and temperature rise rate respectively, the oxygen decoupling temperature analysis of the materials is obtained. Based on the classification of the initial oxygen decoupling temperature, less than 750 °C, 750 - 950 °C, and greater than 950 °C correspond to CLOU, TCES, and TCWS respectively. Finally, the identification results of the material application scenarios are shown in Table 1. The identification accuracy rate of the CLOU scenario is 96.55%, the identification accuracy rate of the TCES scenario is 67.57%, and the identification accuracy rate of the TCWS scenario is 96.39%.

[0078] Table 1 Identification Results of Material Application Scenarios

[0079]

[0080] The above description of the embodiments is to enable those of ordinary skill in the art to understand and use the invention. It is obvious that those skilled in the art can easily make various modifications to these embodiments and apply the general principles described herein to other embodiments without creative efforts. Therefore, the present invention is not limited to the above embodiments, and the improvements and modifications made by those skilled in the art without departing from the scope of the present invention should be within the protection scope of the present invention.

Claims

1. A data-driven rapid prediction method for the thermal weight loss properties of chemical chain oxygen carrier materials, characterized in that: The following steps are involved: S1: Using X-ray diffraction data as the characteristic descriptor of the chemical chain oxygen carrier material structure, using the differential thermogravimetric curve and thermal weight loss properties as the characteristic descriptors of the chemical chain oxygen carrier material performance, using the characteristic descriptor of the material structure and the temperature rise rate as input, and using the characteristic descriptor of the material performance as output, a machine learning model is constructed, and through training, a differential thermogravimetric curve prediction model and a direct prediction model of thermal weight loss properties for the chemical chain oxygen carrier material are obtained respectively; S2: Based on the characteristics of the target material structure and the temperature rise rate, on the one hand, the material derivative thermogravimetric curve is obtained using the trained derivative thermogravimetric curve prediction model, and based on the predicted material derivative thermogravimetric curve, the derivative thermogravimetric is converted into thermogravimetric through integration, and the derivative thermogravimetric and thermogravimetric curves are analyzed to achieve the analysis and prediction of the thermal weight loss properties of the chemical chain oxygen carrier material. On the other hand, the trained direct prediction model for thermal weight loss properties is used to directly predict the thermal weight loss properties of the chemical chain oxygen carrier material.

2. According to claim 1, a data-driven rapid prediction method for thermal weight loss properties of chemical chain oxygen carrier materials is characterized in that: In S1, the X-ray diffraction data and the derivative thermogravimetric curve include: For X-ray diffraction data, the range of diffraction angles was unified to 10-85°, with an angle step of 0.05°, and the range of diffraction intensity data was unified to 0-100, with a total of 1500 sets of data points with one-to-one correspondence between angles and intensities, thus achieving structural unification of X-ray diffraction data; For the Weishang thermogravimetric curve, the temperature range was unified to 700-1000 °C, with a total of 30 sets of data points with one-to-one correspondence between temperature and mass change rate.

3. According to a data-driven rapid prediction method for thermal weight loss properties of chemical chain oxygen carrier materials according to claim 1, it is characterized in that: In S1, the thermal weight loss properties include: Oxygen decoupling temperature and oxygen decoupling mass fraction, wherein the oxygen decoupling temperature includes the initial decoupling temperature and the optimal decoupling temperature.

4. According to a data-driven rapid prediction method for thermal weight loss properties of chemical chain oxygen carrier materials according to claim 1, it is characterized in that: In S1, the process of building a machine learning model includes: Machine learning models are constructed respectively to realize the prediction of the derivative thermogravimetric curve and the direct prediction of the thermal weight loss properties of the chemical chain oxygen carrier material. The machine learning model is used to mine the mapping relationship between the X-ray diffraction data, the temperature rise rate and the derivative thermogravimetric curve and the thermal weight loss properties of the oxygen carrier material.

5. The method for rapid prediction of thermal weight loss properties of chemical chain oxygen carrier materials based on data-driven according to claim 4, characterized in that: In S1, the machine learning model includes one of a convolutional neural network, an artificial neural network, and a Transformer.

6. A data-driven rapid prediction method for thermal weight loss properties of chemical chain oxygen carrier materials according to claim 5, characterized in that: The construction process of the Transformer model includes: The X-ray diffraction data is divided into 30 equal parts according to the diffraction angle, each containing 50 groups of data points, and the corresponding diffraction intensity information is converted into a 30×50 sequence format, with 30 as the time step and 50 as the feature dimension, to match the input format of the Transformer model; After the X-ray diffraction data is embedded in the linear layer, position encoding based on sine-cosine functions is added to add position information to the sequence data, and two layers of Transformer Encoder layers are used as encoders for encoding; Based on the encoded data, the input of temperature rise rate information is added, and then a fully connected layer is used as a decoder for decoding. Finally, it is mapped to the data points of the differential thermogravimetric curve in the form of multiple outputs and mapped to the thermogravimetric properties in the form of a single output.

7. The method for rapid prediction of thermal weight loss properties of chemical chain oxygen carrier materials based on data-driven according to claim 1, characterized in that: In S2, the analysis and prediction of thermogravimetric properties based on differential thermogravimetry and thermogravimetry includes: The data point where the mass change rate of the differential thermogravimetric curve increases significantly is defined as the starting oxygen decoupling point, and its corresponding temperature is the starting oxygen decoupling temperature; the highest point of the differential thermogravimetric curve is defined as the optimal oxygen decoupling point, and its corresponding temperature is the optimal oxygen decoupling temperature; The thermogravimetric curve was obtained by integrating the differential thermogravimetric curve, and the difference in mass percentage corresponding to the initial oxygen decoupling temperature and the final temperature was recorded as the oxygen decoupling mass fraction.

8. A data-driven rapid prediction system for the thermal weight loss properties of chemical chain oxygen carrier materials, characterized in that: include: The model building module uses X-ray diffraction data as the characteristic descriptor of material structure, the differential thermogravimetric curve and thermal weight loss properties as the characteristic descriptors of material performance, and uses structural characteristics and temperature rise rate as input and performance characteristics as output to build a machine learning model to respectively realize the prediction of the differential thermogravimetric curve and the direct prediction of the thermal weight loss properties of chemical chain oxygen carrier materials; The performance prediction module, on the one hand, inputs the material's X-ray diffraction data to obtain the prediction of the material's derivative thermogravimetric curve, obtains the thermogravimetric curve based on the derivative thermogravimetric, and then analyzes and obtains the material's oxygen decoupling temperature and oxygen decoupling mass fraction, to achieve rapid analysis and prediction of the thermal gravimetric properties of the chemical chain oxygen carrier material; on the other hand, inputs the material's structural data to directly obtain the prediction of the thermal gravimetric property data.

9. A storage medium containing computer executable instructions, characterized in that: The storage medium of the computer executable instructions is used to execute the data-driven rapid prediction method for thermal weight loss properties of chemical chain oxygen carrier materials as described in any one of claims 1 to 7 when executed by a computer processor.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, it implements the data-driven rapid prediction method for thermal weight loss properties of chemical chain oxygen carrier materials as described in any one of claims 1 to 7.

Images