Catalyst synthesis parameter optimization method and system based on denitration index evaluation

Abstract

The invention relates to the technical field of catalyst synthesis, and discloses a catalyst synthesis parameter optimization method and system based on denitration index evaluation, and the method comprises the steps: setting a to-be-selected metal element set, obtaining a test catalyst set based on the to-be-selected metal element set, carrying out the catalytic simulation of the test catalyst set according to a plurality of to-be-denitrated scenes and a denitration performance evaluation standard, and obtaining a catalyst denitration data set, performing deep learning by using the catalyst denitration data set to obtain a catalyst component selection model, and obtaining a target catalyst according to the target scene feature vector and the catalyst component selection model. The design accuracy and denitration performance of the catalyst can be improved, and the development period is remarkably shortened.

Description

Technical Field

[0001] This invention relates to the field of catalyst synthesis technology, and in particular to a method and system for optimizing catalyst synthesis parameters based on denitrification index evaluation. Background Technology

[0002] In the field of industrial flue gas treatment, the synthesis of denitrification catalysts is a core step in achieving efficient purification of nitrogen oxides and meeting increasingly stringent environmental emission standards. Developing high-performance and cost-effective catalysts is of paramount importance for the green and sustainable development of highly polluting industries such as waste incineration and steel metallurgy. The precise optimization of synthesis parameters directly determines the denitrification efficiency and lifespan of the catalyst.

[0003] Traditional catalyst synthesis optimization mainly relies on researchers' experience to conduct repeated trial and error experiments. However, the experimental process is highly blind, has a long cycle and high cost. It is also difficult to systematically examine the comprehensive impact of multiple metal elements, raw material ratios and complex operating conditions on performance, and it is easy to miss the optimal solution. At the same time, the optimization dimension of traditional methods is singular, usually only focusing on the denitrification rate, while ignoring key indicators such as catalyst lifetime, synthesis energy consumption and by-product control. Summary of the Invention

[0004] This invention provides a method and system for optimizing catalyst synthesis parameters based on denitrification index evaluation. Its main purpose is to improve the accuracy of catalyst design and denitrification performance, and significantly reduce the development cycle.

[0005] To achieve the above objectives, this invention provides a method for optimizing catalyst synthesis parameters based on denitrification index evaluation, comprising:

[0006] The catalyst product to be synthesized and the low-titanium blast furnace slag were identified, and a set of candidate metal elements was set based on the catalyst product to be synthesized and the low-titanium blast furnace slag.

[0007] The candidate metal elements are extracted sequentially from the candidate metal element set. Based on the candidate metal elements, multiple available metal raw materials are queried. Catalysts are synthesized according to the multiple available metal raw materials to obtain multiple test catalyst groups. The multiple test catalyst groups corresponding to the candidate metal elements are merged to obtain the test catalyst set.

[0008] Multiple denitrification scenarios and denitrification performance evaluation standards were identified. Based on these scenarios and standards, catalytic simulations were performed on the test catalyst set to obtain a catalyst denitrification dataset.

[0009] Deep learning was performed using a catalyst denitrification dataset to obtain a catalyst component selection model;

[0010] Based on the preset target denitrification scenario, a target scenario feature vector is constructed. The target scenario feature vector is then input into the catalyst component selection model to obtain multiple predicted catalyst structure vectors and multiple predicted catalyst performance vectors.

[0011] The target catalyst is selected based on multiple predicted catalyst structure vectors and multiple predicted catalyst performance vectors, and the catalyst synthesis parameters are optimized based on the denitrification index evaluation based on the target catalyst.

[0012] Optionally, the catalyst synthesis is performed based on multiple available metal raw materials to obtain multiple test catalyst groups, including:

[0013] Available metal materials are extracted sequentially from multiple available metal materials, and the following operations are performed on the extracted available metal materials:

[0014] The theoretical proportions of available metal raw materials are set to obtain the catalyst raw material proportion set.

[0015] Catalysts were synthesized according to the catalyst raw material ratio group to obtain the test catalyst group, wherein the proportion of metal raw materials in the catalyst raw material ratio group corresponds one-to-one with the test catalyst in the test catalyst group.

[0016] By summarizing the test catalyst groups corresponding to each available metal raw material from multiple available metal raw materials, multiple test catalyst groups are obtained.

[0017] Optionally, the step of performing catalytic simulation on the test catalyst set according to multiple denitrification scenarios and denitrification performance evaluation standards to obtain a catalyst denitrification dataset includes:

[0018] Test catalysts were extracted sequentially from the test catalyst set, and the following operations were performed on the extracted test catalysts:

[0019] Query the catalyst synthesis energy consumption and catalyst synthesis efficiency during the synthesis process of the tested catalyst;

[0020] Structural features were extracted from the test catalyst to obtain the simulated catalyst structure vector;

[0021] Based on multiple denitrification scenarios, denitrification performance evaluation standards, catalyst synthesis energy consumption, catalyst synthesis efficiency, and simulated catalyst structure vector, multi-scenario denitrification simulations were performed on the test catalysts to obtain multiple catalyst denitrification data.

[0022] By merging multiple catalyst denitrification data corresponding to each test catalyst in the test catalyst set, a catalyst denitrification dataset is obtained.

[0023] Optionally, the step of extracting structural features from the test catalyst to obtain a simulated catalyst structure vector includes:

[0024] Microscopic images of the test catalyst were obtained by taking microscopic photographs.

[0025] The catalyst microscopic image is preprocessed to obtain an enhanced microscopic image, wherein the preprocessing includes: noise reduction and contrast enhancement;

[0026] The enhanced microscopic image is segmented to obtain a binary microscopic image;

[0027] The binary microscopic image is input into a pre-trained feature extractor to obtain a simulated catalyst structure vector, wherein the feature extractor is a convolutional neural network.

[0028] Optionally, based on multiple denitrification scenarios, denitrification performance evaluation standards, catalyst synthesis energy consumption, catalyst synthesis efficiency, and simulated catalyst structure vectors, the test catalyst is subjected to multi-scenario denitrification simulation to obtain multiple catalyst denitrification data, including:

[0029] Extract the denitrification scenarios sequentially from multiple scenarios requiring denitrification;

[0030] Based on the denitrification scenario and the tested catalyst, scenario denitrification simulation was performed to obtain the simulation scenario feature vector and the simulation denitrification performance vector.

[0031] The catalyst synthesis energy consumption and catalyst synthesis efficiency are added to the simulated denitrification performance vector to obtain the simulated catalyst performance vector.

[0032] By merging the simulated catalyst structure vector, the simulated catalyst performance vector, and the simulated scenario feature vector, the original denitrification data is obtained.

[0033] The validity of the original denitrification data is evaluated based on the denitrification performance evaluation standard to obtain the data validity result, where the data validity result is either valid or invalid;

[0034] If the data result is valid, then the original denitrification data shall be recorded as catalyst denitrification data;

[0035] By summarizing the catalyst denitrification data corresponding to the denitrification scenarios, multiple catalyst denitrification data points are obtained.

[0036] Optionally, the step of performing scenario denitrification simulation based on the scenario to be denitrified and the test catalyst to obtain the simulated scenario feature vector and the simulated denitrification performance vector includes:

[0037] Based on the scenario to be denitrified, the composition of waste incineration flue gas was simulated to obtain simulated incineration flue gas. The scenario features of the simulated incineration flue gas were constructed to obtain the simulated scenario feature vector.

[0038] The test catalyst was placed in a pre-constructed reaction tube to obtain a denitrification environment, which included a gas inlet and a gas outlet.

[0039] Denitrification tests were conducted using a denitrification environment and simulated incineration flue gas to obtain the effective denitrification temperature range and effective denitrification rate.

[0040] Record the amount of byproducts generated and the harmfulness index of byproducts in the denitrification test;

[0041] The stable temperature range of the test catalyst is obtained, and the catalyst lifetime is tested based on the stable temperature range and the preset denitrification rate threshold to obtain the catalyst lifetime.

[0042] Vectors are generated based on the effective denitrification temperature range, effective denitrification rate, byproduct generation amount, byproduct harmfulness index, and catalyst lifetime to obtain the simulated denitrification performance vector.

[0043] Optionally, the step of conducting denitrification tests using a denitrification environment and simulated incineration flue gas to obtain the effective denitrification temperature range and effective denitrification rate includes:

[0044] Obtain the catalyst mass in the denitrification environment;

[0045] Based on catalyst quality, denitrification environment, and simulated incineration flue gas, a set of catalyst ambient temperature and an initial denitrification rate set were obtained.

[0046] Based on the catalyst ambient temperature set and the initial denitrification rate set, curve fitting was performed to obtain the denitrification rate-temperature curve;

[0047] Based on the denitrification rate threshold, the effective denitrification temperature range and effective denitrification rate are identified in the denitrification rate temperature curve.

[0048] Optionally, the acquisition of the catalyst ambient temperature set and initial denitrification rate set based on catalyst quality, denitrification environment, and simulated incineration flue gas includes:

[0049] Based on the preset gas volume flow rate and the preset catalyst ambient temperature, the simulated combustion flue gas is introduced into the denitrification environment, and the gas inlet and gas outlet are continuously detected using a pre-built gas analyzer to obtain the inlet nitrification concentration sequence and the outlet nitrification concentration sequence.

[0050] The initial denitrification rate was calculated based on the gas volumetric flow rate, catalyst mass, inlet nitrification concentration sequence, and outlet nitrification concentration sequence.

[0051] The catalyst ambient temperature is adjusted according to a preset temperature change strategy to obtain the adjusted ambient temperature.

[0052] The ambient temperature is adjusted to be the catalyst ambient temperature, and the process of introducing simulated combustion flue gas into the denitrification environment according to the preset gas volume flow rate and the preset catalyst ambient temperature is repeated until a pre-constructed stop test command is received.

[0053] The catalyst ambient temperature and initial denitrification rate are summarized separately to obtain the catalyst ambient temperature set and the initial denitrification rate set.

[0054] Optionally, the step of selecting the target catalyst based on multiple predicted catalyst structure vectors and multiple predicted catalyst performance vectors includes:

[0055] Based on multiple predicted catalyst structure vectors, identify multiple similar catalysts in a pre-defined catalyst structure feature table;

[0056] Receive a set of catalyst performance weights, wherein the number of catalyst performance weights in the set is the same as the vector dimension of the predicted catalyst performance vector;

[0057] The predicted catalyst performance vectors are extracted sequentially from multiple predicted catalyst performance vectors, and a weighted calculation is performed based on the predicted catalyst performance vectors and the catalyst performance weight set to obtain the comprehensive catalytic performance value.

[0058] By summing up the comprehensive catalytic performance values, multiple comprehensive catalytic performance values ​​are obtained, and the maximum catalytic performance value among the multiple comprehensive catalytic performance values ​​is identified.

[0059] Identify the target catalyst with the highest catalytic performance value among multiple similar catalysts.

[0060] To achieve the above objectives, the present invention also provides a catalyst synthesis parameter optimization system based on denitrification index evaluation, comprising:

[0061] The metal element setting module is used to identify the catalyst product to be synthesized and the low-titanium blast furnace slag, and to set a set of candidate metal elements based on the catalyst product to be synthesized and the low-titanium blast furnace slag.

[0062] The catalyst preparation module is used to sequentially extract candidate metal elements from the candidate metal element set, query multiple available metal raw materials based on the candidate metal elements, synthesize catalysts according to the multiple available metal raw materials, obtain multiple test catalyst groups, and merge the multiple test catalyst groups corresponding to the candidate metal elements to obtain the test catalyst set.

[0063] A model building module was selected to identify multiple denitrification scenarios and denitrification performance evaluation standards. Based on these scenarios and standards, catalytic simulations were performed on the test catalyst set to obtain a catalyst denitrification dataset. Deep learning was then performed on this dataset to obtain a catalyst component selection model.

[0064] The target catalyst selection module is used to construct a target scenario feature vector based on a preset target denitrification scenario, input the target scenario feature vector into the catalyst component selection model, obtain multiple predicted catalyst structure vectors and multiple predicted catalyst performance vectors, and select the target catalyst based on the multiple predicted catalyst structure vectors and multiple predicted catalyst performance vectors.

[0065] To address the above problems, the present invention also provides an electronic device, the electronic device comprising:

[0066] Memory, storing at least one instruction;

[0067] The processor executes the instructions stored in the memory to implement the catalyst synthesis parameter optimization method based on denitrification index evaluation described above.

[0068] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the catalyst synthesis parameter optimization method based on denitrification index evaluation described above.

[0069] To address the problems described in the background art, this invention first synthesizes and summarizes catalysts from multiple available metal raw materials to obtain a test catalyst set. This step systematically explores various available chemical raw materials (such as nitrates, sulfates, etc.) and different theoretical ratios of active components for each candidate metal element, thereby constructing a test catalyst set with highly diverse composition and structure. This greatly expands the range of candidate catalysts and avoids the problem of missing optimal solutions due to the selection of only one raw material and ratio. Then, catalytic simulation is performed on the test catalyst set according to multiple denitrification scenarios and denitrification performance evaluation standards to obtain a catalyst denitrification dataset. This step introduces a comprehensive evaluation system. This step not only simulates various actual operating conditions (such as different temperatures and impurity gas concentrations) to test the adaptability of the catalyst, but also introduces a digital vector of the catalyst microstructure (simulated catalyst structure vector) and energy consumption and efficiency indicators of the synthesis process, so that the final dataset can comprehensively reflect the structure and performance of the catalyst. The complex relationship between synthesis economy and environmental adaptability provides high-quality multi-dimensional features for model learning. Furthermore, deep learning is performed using a catalyst denitrification dataset to obtain a catalyst component selection model. This step utilizes deep learning technology to automatically mine the deep nonlinear mapping relationship between the catalyst's composition, structure, synthesis parameters, and performance under various scenarios from massive, high-dimensional experimental data. This establishes an intelligent prediction model that can replace numerous trial-and-error experiments. Finally, based on the target scenario feature vector and the catalyst component selection model, the target catalyst is obtained. This step achieves customized and forward-looking prediction of catalyst design. Users only need to input the target operating conditions, and the model can directly output a series of predicted structural features and comprehensive performance indicators of potentially applicable catalysts. This allows for rapid screening of potential candidate solutions before actual synthesis, greatly shortening the R&D cycle. Therefore, this invention can improve the accuracy of catalyst design and denitrification performance, and significantly reduce the development cycle. Attached Figure Description

[0070] Figure 1 This is a schematic flowchart of a catalyst synthesis parameter optimization method based on denitrification index evaluation provided in an embodiment of the present invention;

[0071] Figure 2 This is a functional block diagram of a catalyst synthesis parameter optimization system based on denitrification index evaluation provided in an embodiment of the present invention;

[0072] Figure 3 This is a schematic diagram of an electronic device for implementing the catalyst synthesis parameter optimization method based on denitrification index evaluation, according to an embodiment of the present invention.

[0073] Explanation of reference numerals in the attached figures:

[0074] 10. Electronic device; 11. Processor; 12. Memory; 13. Bus.

[0075] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0076] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0077] This application provides a method for optimizing catalyst synthesis parameters based on denitrification index evaluation. The execution entity of this method includes, but is not limited to, at least one electronic device that can be configured to execute the method provided in this application, such as a server or a terminal. In other words, the method can be executed by software or hardware installed on a terminal device or server device, and the software may be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0078] Reference Figure 1 The diagram shown is a schematic flowchart of a catalyst synthesis parameter optimization method based on denitrification index evaluation according to an embodiment of the present invention. In this embodiment, the catalyst synthesis parameter optimization method based on denitrification index evaluation includes:

[0079] S1. Identify the catalyst product to be synthesized and the low-titanium blast furnace slag, and set a set of candidate metal elements based on the catalyst product to be synthesized and the low-titanium blast furnace slag.

[0080] It is clear that the catalyst product to be synthesized refers to the product that needs to be synthesized. In this embodiment, the catalyst product to be synthesized is: MeOx-MnO2-CeO2, where Me in MeOx represents a metal atom, O represents an oxygen atom, x represents the number of oxygen atoms in MeOx, MnO2 represents manganese dioxide as one of the main active components, which is used to provide good redox ability, and CeO2 represents cerium dioxide as a key active component and structure promoter, which can significantly broaden the active temperature window of the catalyst and improve its stability. The low-titanium blast furnace slag refers to a metallurgical industrial by-product containing a small amount of titanium dioxide. In this method, it is mainly used as a catalyst support or substrate. Its porous structure, mechanical strength, and certain beneficial elements (such as titanium) may be used to reduce the cost of the catalyst and optimize its physical structure. The set of candidate metal elements refers to a collection of multiple candidate metal elements. Among them, the candidate metal elements refer to the third metal elements that are planned to be doped into the MnO2-CeO2 basic framework to modulate its electronic structure, surface acidity, or redox properties in order to further improve the denitrification performance. Examples include: iron (Fe), copper (Cu), cobalt (Co), tungsten (W), vanadium (V), etc. The selection of candidate metal elements based on the catalyst product to be synthesized and low-titanium blast furnace slag refers to the following: metal elements selected by relevant experts, taking into account prior knowledge (such as metals known in literature to promote the denitrification reaction) and the chemical compatibility of low-titanium blast furnace slag (avoiding the selection of metals that will react adversely with the components in the slag). For example, Fe and Cu are common active components for denitrification, and their oxides have good chemical compatibility with low-titanium blast furnace slag at high temperatures, so they are included in the selection of candidate metal elements.

[0081] S2. Extract the candidate metal elements sequentially from the candidate metal element set, query multiple available metal raw materials based on the candidate metal elements, synthesize catalysts according to the multiple available metal raw materials, obtain multiple test catalyst groups, merge the multiple test catalyst groups corresponding to the candidate metal elements, and obtain the test catalyst set.

[0082] It should be explained that the "available metal raw materials" refer to common chemical reagents corresponding to the candidate metal element that can be used to synthesize catalysts under laboratory conditions. These available metal raw materials are solid powders or crystals, and are often prepared into aqueous solutions for use in the laboratory. For example, if the candidate metal element is iron (Fe), then the available metal raw materials for this candidate metal element can be: ferric nitrate, ferric sulfate, ferric chloride, ferric acetate, etc. The "test catalyst set" refers to a collection of multiple catalysts prepared from available metal raw materials, wherein one available metal raw material corresponds to one test catalyst set. The "merging of multiple test catalyst sets corresponding to the candidate metal element" means: putting the test catalysts from all the multiple test catalyst sets corresponding to the candidate metal element into the same data combination; the resulting data combination is the test catalyst set.

[0083] In detail, the catalyst is synthesized based on multiple available metal raw materials to obtain multiple test catalyst groups, including:

[0084] Available metal materials are extracted sequentially from multiple available metal materials, and the following operations are performed on the extracted available metal materials:

[0085] The theoretical proportions of available metal raw materials are set to obtain the catalyst raw material proportion set.

[0086] Catalysts were synthesized according to the catalyst raw material ratio group to obtain the test catalyst group, wherein the proportion of metal raw materials in the catalyst raw material ratio group corresponds one-to-one with the test catalyst in the test catalyst group.

[0087] By summarizing the test catalyst groups corresponding to each available metal raw material from multiple available metal raw materials, multiple test catalyst groups are obtained.

[0088] Understandably, the theoretical ratio setting refers to a series of molar ratios or mass ratios between artificially pre-designed active components (i.e., MeOx, MnO2 and CeO2), for example: setting three molar ratios: Me:Mn:Ce=1:1:1, 1:2:1 and 2:1:1. The catalyst raw material ratio group refers to a collection of multiple catalyst raw material ratios. The catalyst raw material ratio refers to the mass ratio of the selected metal raw material to the raw material for preparing MnO2 (manganese nitrate) and the raw material for preparing CeO2 (cerium nitrate), obtained after a set procedure. For example, a catalyst raw material ratio group might be: {Ratio A: Copper nitrate: Manganese nitrate: Cerium nitrate = 2.41g:1.79g:3.25g (corresponding to a Me:Mn:Ce molar ratio of 1:1:1), Ratio B: Copper nitrate: Manganese nitrate: Cerium nitrate = 2.41g:3.58g:3.25g (corresponding to a molar ratio of 1:2:1), Ratio C: Copper nitrate: Manganese nitrate: Cerium nitrate = 4.82g:1.79g:3.25g (corresponding to a molar ratio of 2:1:1)}. The test catalyst group refers to a collection of multiple test catalysts. The test catalyst refers to a catalyst prepared according to the catalyst raw material ratio, and the preparation method includes co-precipitation, impregnation, etc.

[0089] S3. Identify multiple denitrification scenarios and denitrification performance evaluation standards. Perform catalytic simulation on the test catalyst set based on the multiple denitrification scenarios and denitrification performance evaluation standards to obtain the catalyst denitrification dataset.

[0090] It is clear that the denitrification scenarios refer to a series of test environments set up to simulate different operating conditions that the catalyst may encounter in actual applications, such as: standard waste incineration flue gas (containing NO 500ppm, SO2 100ppm, H2O 10%, 350°C), high water vapor scenarios (H2O content increased to 15%), low temperature scenarios (temperature reduced to 250°C), and harsh scenarios containing high concentrations of SO2 (SO2 increased to 200ppm), etc. The denitrification performance evaluation standard refers to an artificially set minimum performance threshold standard used to screen qualified catalysts. For example, this denitrification performance evaluation standard sets the minimum denitrification rate to be greater than 3.5 × 10⁻⁻⁻⁻⁶. 7 The minimum value for byproduct formation is set to ≤15 ppm (mol / (g·s) at a standard 350°C setting). The catalyst denitrification dataset refers to a collection of data obtained through catalytic simulation, containing detailed performance indicators (such as denitrification rate, byproduct formation, etc.) for each tested catalyst under all denitrification scenarios.

[0091] In detail, the catalyst set is subjected to catalytic simulation based on multiple denitrification scenarios and denitrification performance evaluation standards to obtain a catalyst denitrification dataset, including:

[0092] Test catalysts were extracted sequentially from the test catalyst set, and the following operations were performed on the extracted test catalysts:

[0093] Query the catalyst synthesis energy consumption and catalyst synthesis efficiency during the synthesis process of the tested catalyst;

[0094] Structural features were extracted from the test catalyst to obtain the simulated catalyst structure vector;

[0095] Based on multiple denitrification scenarios, denitrification performance evaluation standards, catalyst synthesis energy consumption, catalyst synthesis efficiency, and simulated catalyst structure vector, multi-scenario denitrification simulations were performed on the test catalysts to obtain multiple catalyst denitrification data.

[0096] By merging multiple catalyst denitrification data corresponding to each test catalyst in the test catalyst set, a catalyst denitrification dataset is obtained.

[0097] It is clear that the catalyst synthesis energy consumption refers to the total energy consumed in the preparation of the test catalyst divided by the total mass of the prepared test catalyst, mainly including the electrical or thermal energy consumption of heat treatment steps such as calcination and drying. The catalyst synthesis efficiency refers to the mass of qualified catalyst successfully synthesized per unit time or the conversion rate of unit raw materials, used to measure the rate of synthesis. The simulated catalyst structure vector refers to a high-dimensional digital vector representing the microscopic structural features of the test catalyst extracted from the catalyst microscopic image through a convolutional neural network. The function of the simulated catalyst structure vector is to quantify the physical morphology of the catalyst (such as particle size, porosity, and dispersion) and use it as a key feature input for subsequent establishment of a catalyst component selection model. The catalyst denitrification data refers to a complete data record obtained after multi-scenario denitrification simulation, containing all performance indicators (such as the effective denitrification temperature range), synthesis parameters (such as catalyst synthesis efficiency), and structural features (simulated catalyst structure vector) of the test catalyst in a certain denitrification scenario. One denitrification scenario corresponds to one catalyst denitrification data.

[0098] In detail, the extraction of structural features from the test catalyst to obtain a simulated catalyst structure vector includes:

[0099] Microscopic images of the test catalyst were obtained by taking microscopic photographs.

[0100] The catalyst microscopic image is preprocessed to obtain an enhanced microscopic image, wherein the preprocessing includes: noise reduction and contrast enhancement;

[0101] The enhanced microscopic image is segmented to obtain a binary microscopic image;

[0102] The binary microscopic image is input into a pre-trained feature extractor to obtain a simulated catalyst structure vector, wherein the feature extractor is a convolutional neural network.

[0103] As is clear, the catalyst microscopic image refers to an image of the test catalyst obtained through microscopic imaging. Microscopic imaging refers to acquiring grayscale images of the catalyst sample surface or internal structure at a microscopic scale (such as nanometer or micrometer scale) using equipment such as scanning electron microscopes (SEM) or transmission electron microscopes (TEM). The enhanced microscopic image refers to a catalyst microscopic image after preprocessing. Noise reduction refers to using algorithms such as Gaussian filtering and median filtering to reduce random pixel noise introduced by the instrument or environment, making the catalyst microscopic image smoother and its feature contours clearer. Contrast enhancement refers to adjusting the grayscale value distribution of the image through methods such as histogram equalization, expanding the difference in brightness between the foreground (particles in the test catalyst) and the background, making structural details easier to identify. The binary microscopic image refers to the segmented enhanced microscopic image. The segmentation method can be selected from global thresholding, adaptive thresholding, edge detection (such as the Canny algorithm), or a deep learning-based semantic segmentation model (such as U-Net). This binary microscopic image can clearly separate the particles or pore regions (foreground) in the test catalyst from the background. The foreground is represented by white (pixel value 255), and the background by black (pixel value 0), laying the foundation for subsequent quantitative analysis. The feature extractor refers to a convolutional neural network (CNN, such as VGG or ResNet) pre-trained on a large-scale image dataset (such as ImageNet). This feature extractor can use convolutional layers to convert the input image into a high-dimensional, low-dimensional numerical vector (i.e., a simulated catalyst structure vector). This vector can efficiently capture complex, abstract feature patterns in the image that are difficult for the human eye to quantify.

[0104] In detail, based on multiple denitrification scenarios, denitrification performance evaluation standards, catalyst synthesis energy consumption, catalyst synthesis efficiency, and simulated catalyst structure vectors, multi-scenario denitrification simulations were performed on the test catalyst to obtain multiple catalyst denitrification data, including:

[0105] Extract the denitrification scenarios sequentially from multiple scenarios requiring denitrification;

[0106] Based on the denitrification scenario and the tested catalyst, scenario denitrification simulation was performed to obtain the simulation scenario feature vector and the simulation denitrification performance vector.

[0107] The catalyst synthesis energy consumption and catalyst synthesis efficiency are added to the simulated denitrification performance vector to obtain the simulated catalyst performance vector.

[0108] By merging the simulated catalyst structure vector, the simulated catalyst performance vector, and the simulated scenario feature vector, the original denitrification data is obtained.

[0109] The validity of the original denitrification data is evaluated based on the denitrification performance evaluation standard to obtain the data validity result, where the data validity result is either valid or invalid;

[0110] If the data result is valid, then the original denitrification data shall be recorded as catalyst denitrification data;

[0111] By summarizing the catalyst denitrification data corresponding to the denitrification scenarios, multiple catalyst denitrification data points are obtained.

[0112] It is clear that the simulated scenario feature vector refers to a vector that quantifies the characteristics of the scenario to be denitrified. This simulated scenario feature vector may include key operating parameters such as temperature, pressure, initial NOx concentration, SO2 concentration, water vapor concentration, and space velocity. The simulated denitrification performance vector refers to a comprehensive vector that quantifies the catalytic performance of the catalyst under the scenario to be denitrified. The simulated catalyst performance vector refers to the simulated denitrification performance vector after adding catalyst synthesis energy consumption and catalyst synthesis efficiency. That is, based on the catalytic performance indicators, indicators characterizing the energy consumption and efficiency of the synthesis process are added to form a more comprehensive performance description. The original denitrification data refers to the set of simulated catalyst structure vector, simulated catalyst performance vector, and simulated scenario feature vector. The valid data result refers to the result obtained after data validity evaluation. The data validity evaluation refers to: determining whether the simulated catalyst performance vector in the original denitrification data meets the denitrification performance evaluation criteria. If it does, the valid data result is recorded as valid; otherwise, it is recorded as invalid.

[0113] For example, if the denitrification performance evaluation standard requires a denitrification rate > 3.5 × 10⁻ 7 The denitrification rate of a certain catalyst is 4.0 × 10⁻⁻ mol / (g·s) and the byproduct formation is ≤15 ppm. 7 If the concentration of byproducts is mol / (g·s) and the amount of byproducts generated is 12 ppm, then the data is considered valid.

[0114] In detail, the scenario-based denitrification simulation based on the denitrification scenario to be denitrified and the test catalyst, to obtain the simulated scenario feature vector and the simulated denitrification performance vector, includes:

[0115] Based on the scenario to be denitrified, the composition of waste incineration flue gas was simulated to obtain simulated incineration flue gas. The scenario features of the simulated incineration flue gas were constructed to obtain the simulated scenario feature vector.

[0116] The test catalyst was placed in a pre-constructed reaction tube to obtain a denitrification environment, which included a gas inlet and a gas outlet.

[0117] Denitrification tests were conducted using a denitrification environment and simulated incineration flue gas to obtain the effective denitrification temperature range and effective denitrification rate.

[0118] Record the amount of byproducts generated and the harmfulness index of byproducts in the denitrification test;

[0119] The stable temperature range of the test catalyst is obtained, and the catalyst lifetime is tested based on the stable temperature range and the preset denitrification rate threshold to obtain the catalyst lifetime.

[0120] Vectors are generated based on the effective denitrification temperature range, effective denitrification rate, byproduct generation amount, byproduct harmfulness index, and catalyst lifetime to obtain the simulated denitrification performance vector.

[0121] It should be explained that the simulated incineration flue gas refers to a mixed gas with specific components and concentrations obtained after simulating the composition of waste incineration flue gas. The simulation of the composition of waste incineration flue gas means that, based on the components of the scenario to be denitrified, high-purity nitrogen is used as the balance gas, and NO, NH3, SO2, water vapor, O2 and other gas components are precisely controlled and mixed into the balance gas through a mass flow meter to simulate the chemical environment of real waste incineration flue gas. The simulated scenario feature vector refers to a vector that quantifies the characteristics of simulated incineration flue gas. Constructing scenario features for simulated incineration flue gas involves combining key operating parameters of the simulated flue gas, such as temperature, pressure, concentrations of various gas components (NO, SO2, H2O, etc.), and space velocity (GHSV), into a one-dimensional array in a predefined order. For example, a simulated scenario feature vector can be represented as [350, 101.325, 500, 100, 10, 5, 50000], corresponding to temperature (°C), pressure (kPa), NO concentration (ppm), SO2 concentration (ppm), H2O concentration (%), O2 concentration (%), and space velocity (h⁻¹), respectively. The reaction tube refers to a quartz tube or stainless steel tube, which is the core device for catalyst performance testing.

[0122] It is clear that the denitrification environment refers to the reaction tube after the test catalyst has been placed inside. The gas inlet and gas outlet refer to the input and output ports of the simulated combustion flue gas, respectively. The effective denitrification temperature range refers to the temperature range within which the test catalyst can operate stably. It represents the effective temperature range (e.g., 250°C to 400°C) that the catalyst can maintain when the denitrification rate is higher than a preset threshold (e.g., 90%), and is a key indicator for evaluating the catalyst's adaptability to operating conditions. The effective denitrification rate refers to the denitrification rate of the test catalyst.

[0123] Furthermore, the byproduct generation amount refers to the concentration of non-target nitrogen oxides (target nitrogen oxides such as N2O and NO2) generated during the denitrification test, and the byproduct hazard index refers to a numerical value (such as global warming potential value) quantifying the environmental harm of non-target nitrogen oxides generated during the denitrification test, which is set manually. The stable temperature range refers to the temperature region in the denitrification rate temperature curve where the denitrification rate is higher than the denitrification rate threshold. The denitrification rate threshold refers to a manually set minimum denitrification rate. When the catalyst's denitrification rate is lower than this threshold, it indicates that the tested catalyst no longer has industrial application value. The catalyst lifetime test based on the stable temperature range and the preset denitrification rate threshold refers to: selecting a representative temperature point (such as the median temperature) within the stable temperature range, continuously introducing simulated combustion flue gas at this temperature and monitoring the change in its denitrification rate over time until the denitrification rate decays below the threshold, and recording the duration of this process as the catalyst lifetime. The vector generation refers to converting the effective denitrification temperature span, effective denitrification rate, byproduct generation amount, byproduct hazard index, and catalyst lifetime into a vector form.

[0124] In detail, the denitrification test using a denitrification environment and simulated incineration flue gas to obtain the effective denitrification temperature range and effective denitrification rate includes:

[0125] Obtain the catalyst mass in the denitrification environment;

[0126] Based on catalyst quality, denitrification environment, and simulated incineration flue gas, a set of catalyst ambient temperature and an initial denitrification rate set were obtained.

[0127] Based on the catalyst ambient temperature set and the initial denitrification rate set, curve fitting was performed to obtain the denitrification rate-temperature curve;

[0128] Based on the denitrification rate threshold, the effective denitrification temperature range and effective denitrification rate are identified in the denitrification rate temperature curve.

[0129] Understandably, the catalyst mass refers to the mass of the catalyst tested in the denitrification environment. The catalyst ambient temperature set refers to the set of denitrification ambient temperatures during the subsequent introduction of simulated incineration flue gas into the denitrification environment. The initial denitrification rate set refers to the set of catalyst catalytic rates tested during the subsequent introduction of simulated incineration flue gas into the denitrification environment. The catalyst ambient temperatures in the catalyst ambient temperature set correspond one-to-one with the initial denitrification rates in the initial denitrification rate set. The denitrification rate-temperature curve refers to a curve representing the relationship between the catalyst ambient temperature and the initial denitrification rate, where the horizontal axis represents the catalyst ambient temperature and the vertical axis represents the initial denitrification rate. The above-mentioned identification of the effective denitrification temperature span and effective denitrification rate in the denitrification rate-temperature curve based on the denitrification rate threshold refers to: identifying the regions in the denitrification rate-temperature curve where the vertical axis is greater than the denitrification rate threshold for the catalyst ambient temperature, and recording the total length of the horizontal axis of these regions (i.e., the total temperature span represented by these regions) as the effective denitrification temperature span, and recording the average of all initial denitrification rates within these regions as the effective denitrification rate. If there are no coordinate points in the denitrification rate-temperature curve higher than the denitrification rate threshold, then both the effective denitrification temperature span and the effective denitrification rate are recorded as 0. For example, the denitrification rate threshold is 2.0 × 10⁻⁻⁻⁶. 7 mol / (g·s), in the denitrification rate temperature curve, is higher than 2.0×10⁻ 7 If the temperature range (i.e., the horizontal axis range) with mol / (g·s) is 200℃ to 300℃, then the effective denitrification temperature range is 100℃, and the average initial denitrification rate within the 200℃ to 300℃ range is 3.5×10⁻ 7 If the effective denitrification rate is 3.5 × 10⁻ mol / (g·s), then the effective denitrification rate is 3.5 × 10⁻ 7 mol / (g·s).

[0130] Specifically, the acquisition of the catalyst ambient temperature set and initial denitrification rate set based on catalyst quality, denitrification environment, and simulated combustion flue gas includes:

[0131] Based on the preset gas volume flow rate and the preset catalyst ambient temperature, the simulated combustion flue gas is introduced into the denitrification environment, and the gas inlet and gas outlet are continuously detected using a pre-built gas analyzer to obtain the inlet nitrification concentration sequence and the outlet nitrification concentration sequence.

[0132] The initial denitrification rate was calculated based on the gas volumetric flow rate, catalyst mass, inlet nitrification concentration sequence, and outlet nitrification concentration sequence.

[0133] The catalyst ambient temperature is adjusted according to a preset temperature change strategy to obtain the adjusted ambient temperature.

[0134] The ambient temperature is adjusted to be the catalyst ambient temperature, and the process of introducing simulated combustion flue gas into the denitrification environment according to the preset gas volume flow rate and the preset catalyst ambient temperature is repeated until a pre-constructed stop test command is received.

[0135] The catalyst ambient temperature and initial denitrification rate are summarized separately to obtain the catalyst ambient temperature set and the initial denitrification rate set.

[0136] It is clear that the gas volumetric flow rate refers to the volume of simulated flue gas introduced into the denitrification environment per unit time, which is set artificially. The catalyst ambient temperature refers to the temperature in the denitrification environment, which will be continuously iterated with subsequent adjustments. The gas analyzer refers to an instrument used for real-time online monitoring of the concentration of specific components in the gas, such as a Fourier transform infrared spectrometer (FTIR). The inlet nitrification concentration sequence refers to the set of multiple nitrification concentrations (at different times) detected by the gas detector at the gas inlet (arranged from front to back according to the acquisition time). The outlet nitrification concentration sequence refers to the set of multiple nitrification concentrations (at different times) detected by the gas detector at the gas outlet. The initial denitrification rate refers to the denitrification rate of the catalyst tested during the acquisition of the inlet and outlet nitrification concentration sequences. The temperature change strategy refers to a pre-set program for systematically changing the reaction temperature, for example: starting from 150°C, the temperature is increased to 450°C at a rate of 5°C / min, and held constant at each temperature point for 10 minutes to collect stable data. The adjusted ambient temperature refers to the catalyst ambient temperature after adjustment. The stop test command refers to a manually set command to stop the introduction of simulated combustion flue gas. For example, when the relevant operator observes that the strategy temperature rise has reached the preset maximum temperature (e.g., 450°C), or the catalyst's denitrification activity has been completely lost (the outlet nitrification concentration is close to the inlet nitrification concentration), a stop test command is issued.

[0137] Furthermore, the above-mentioned catalyst denitration rate is calculated as follows: ,in, Indicates the catalyst denitration rate. This represents the average value of all inlet nitrate concentrations in the inlet nitrate concentration sequence. This represents the average value of all outlet nitrate concentrations in the outlet nitrate concentration sequence. Indicates gas volume flow rate, This indicates the mass of the catalyst.

[0138] S4. Deep learning is performed using the catalyst denitrification dataset to obtain a catalyst component selection model.

[0139] As is clear, the catalyst cost selection model refers to a neural network model that can automatically select the optimal catalyst composition and performance parameters according to different denitrification scenarios. This neural network model can be a deep fully connected neural network or a Transformer encoder specifically designed for processing structured data, etc. The input of this catalyst cost selection model is the scenario feature vector under a certain denitrification scenario, and the output is the structure vector (corresponding to the simulated catalyst structure vector) and performance vector (corresponding to the simulated catalyst performance vector) of a certain catalyst. It should be noted that before using the catalyst denitrification dataset for deep learning, preprocessing steps such as normalization, data cleaning (e.g., removing invalid data) and feature scaling are required. These preprocessing steps will not be elaborated here.

[0140] S5. Construct a target scenario feature vector based on the preset target denitrification scenario, input the target scenario feature vector into the catalyst component selection model, and obtain multiple predicted catalyst structure vectors and multiple predicted catalyst performance vectors.

[0141] Understandably, the target denitrification scenario refers to the specific scenario where denitrification needs to be performed. The target scenario feature vector refers to a vector quantifying the characteristics of the target denitrification scenario; the construction method of this target scenario feature vector is the same as that of the simulated scenario feature vector, and will not be elaborated further here. The multiple predicted catalyst structure vectors refer to the vectors of multiple catalyst structures (corresponding to the simulated catalyst structure vectors) output by the catalyst cost selection model. The multiple predicted catalyst performance vectors refer to the vectors of multiple catalyst performances (corresponding to the simulated catalyst performance vectors) output by the catalyst selection model, wherein there is a one-to-one correspondence between the predicted catalyst structure vectors and the predicted catalyst performance vectors.

[0142] S6. Select the target catalyst based on multiple predicted catalyst structure vectors and multiple predicted catalyst performance vectors, and optimize the catalyst synthesis parameters based on the denitrification index evaluation based on the target catalyst.

[0143] It needs to be explained that different denitrification process standards have different requirements for denitrification performance. For example, in areas with stricter emission standards, the denitrification rate and the harmful index of by-products may be the primary considerations, and the catalyst performance weights corresponding to them are the highest. However, for projects that need to control operating energy consumption, the catalyst performance weights corresponding to catalyst lifetime and synthesis energy consumption may be set higher. Therefore, it is necessary to select from multiple predicted catalyst structure vectors output by the catalyst component selection model according to different denitrification performance requirements (hence the introduction of catalyst performance weights to quantify these requirements) in order to obtain the catalyst that best meets the current denitrification performance requirements.

[0144] Specifically, the step of selecting the target catalyst based on multiple predicted catalyst structure vectors and multiple predicted catalyst performance vectors includes:

[0145] Based on multiple predicted catalyst structure vectors, identify multiple similar catalysts in a pre-defined catalyst structure feature table;

[0146] Receive a set of catalyst performance weights, wherein the number of catalyst performance weights in the set is the same as the vector dimension of the predicted catalyst performance vector;

[0147] The predicted catalyst performance vectors are extracted sequentially from multiple predicted catalyst performance vectors, and a weighted calculation is performed based on the predicted catalyst performance vectors and the catalyst performance weight set to obtain the comprehensive catalytic performance value.

[0148] By summing up the comprehensive catalytic performance values, multiple comprehensive catalytic performance values ​​are obtained, and the maximum catalytic performance value among the multiple comprehensive catalytic performance values ​​is identified.

[0149] Identify the target catalyst with the highest catalytic performance value among multiple similar catalysts.

[0150] It should be explained that the catalyst structure feature table refers to an artificially constructed database containing the mapping relationship between known catalyst samples and their digitized structural features. For example, the test catalyst from step S2 above and the corresponding simulated catalyst structure vector can be included in this catalyst structure feature table (corresponding to the catalyst and structural feature vector in the catalyst structure feature table, respectively). The similar catalyst refers to the catalyst corresponding to the structural feature vector in the catalyst structure feature table that has a high similarity to the predicted catalyst structure vector. The identification method is as follows: extract the predicted catalyst structure vector sequentially from multiple predicted catalyst structure vectors, calculate the cosine similarity between the predicted catalyst structure vector and each structural feature vector in the catalyst structure feature table, obtain multiple cosine similarities, record the structural feature vector corresponding to the maximum similarity among the multiple cosine similarities as the similar feature vector, record the catalyst corresponding to the similar feature vector as the similar catalyst, and summarize the similar catalysts to obtain multiple similar catalysts.

[0151] Furthermore, the catalytic performance weight set refers to a set of importance weights artificially assigned to each performance indicator (such as denitrification rate, catalyst lifetime, synthesis energy consumption, etc.) in the predicted catalyst performance vector according to actual project requirements (such as focusing on efficiency, lifetime, or energy consumption). The sum of all catalytic performance weights in the weight set is 1. For example, if the performance vector includes three indicators: [effective denitrification rate, catalyst lifetime, catalyst synthesis energy consumption], the weight set can be set to [0.5, 0.3, 0.2], indicating that denitrification efficiency is the most important, followed by catalyst lifetime, and finally catalyst synthesis energy consumption. The comprehensive catalytic performance value refers to a single scalar value that quantifies and predicts the overall performance of the catalyst. The weighted calculation refers to multiplying each catalyst performance weight in the weight set with each vector element in the predicted catalyst performance vector and then summing the results.

[0152] To address the problems described in the background art, this invention first synthesizes and summarizes catalysts from multiple available metal raw materials to obtain a test catalyst set. This step systematically explores various available chemical raw materials (such as nitrates, sulfates, etc.) and different theoretical ratios of active components for each candidate metal element, thereby constructing a test catalyst set with highly diverse composition and structure. This greatly expands the range of candidate catalysts and avoids the problem of missing optimal solutions due to the selection of only one raw material and ratio. Then, catalytic simulation is performed on the test catalyst set according to multiple denitrification scenarios and denitrification performance evaluation standards to obtain a catalyst denitrification dataset. This step introduces a comprehensive evaluation system. This step not only simulates various actual operating conditions (such as different temperatures and impurity gas concentrations) to test the adaptability of the catalyst, but also introduces a digital vector of the catalyst microstructure (simulated catalyst structure vector) and energy consumption and efficiency indicators of the synthesis process, so that the final dataset can comprehensively reflect the structure and performance of the catalyst. The complex relationship between synthesis economy and environmental adaptability provides high-quality multi-dimensional features for model learning. Furthermore, deep learning is performed using a catalyst denitrification dataset to obtain a catalyst component selection model. This step utilizes deep learning technology to automatically mine the deep nonlinear mapping relationship between the catalyst's composition, structure, synthesis parameters, and performance under various scenarios from massive, high-dimensional experimental data. This establishes an intelligent prediction model that can replace numerous trial-and-error experiments. Finally, based on the target scenario feature vector and the catalyst component selection model, the target catalyst is obtained. This step achieves customized and forward-looking prediction of catalyst design. Users only need to input the target operating conditions, and the model can directly output a series of predicted structural features and comprehensive performance indicators of potentially applicable catalysts. This allows for rapid screening of potential candidate solutions before actual synthesis, greatly shortening the R&D cycle. Therefore, this invention can improve the accuracy of catalyst design and denitrification performance, and significantly reduce the development cycle.

[0153] like Figure 2 The diagram shown is a functional block diagram of a catalyst synthesis parameter optimization system based on denitrification index evaluation provided in an embodiment of the present invention.

[0154] The catalyst synthesis parameter optimization system 100 based on denitrification index evaluation described in this invention can be installed in an electronic device. Depending on the functions implemented, the catalyst synthesis parameter optimization system 100 may include a metal element setting module 101, a catalyst preparation test module 102, a model selection and construction module 103, and a target catalyst selection module 104. The module described in this invention can also be called a unit, referring to a series of computer program segments that can be executed by an electronic device processor and perform a fixed function, stored in the electronic device's memory.

[0155] The metal element setting module 101 is used to identify the catalyst product to be synthesized and the low-titanium blast furnace slag, and set a set of candidate metal elements based on the catalyst product to be synthesized and the low-titanium blast furnace slag.

[0156] The test catalyst preparation module 102 is used to sequentially extract the candidate metal elements from the candidate metal element set, query multiple available metal raw materials based on the candidate metal elements, synthesize catalysts according to the multiple available metal raw materials, obtain multiple test catalyst groups, and merge the multiple test catalyst groups corresponding to the candidate metal elements to obtain a test catalyst set.

[0157] The selected model construction module 103 is used to identify multiple denitrification scenarios and denitrification performance evaluation standards, perform catalytic simulation on the test catalyst set according to the multiple denitrification scenarios and denitrification performance evaluation standards, obtain a catalyst denitrification dataset, and use the catalyst denitrification dataset for deep learning to obtain a catalyst component selection model.

[0158] The target catalyst selection module 104 is used to construct a target scenario feature vector based on a preset target denitrification scenario, input the target scenario feature vector into the catalyst component selection model to obtain multiple predicted catalyst structure vectors and multiple predicted catalyst performance vectors, and select a target catalyst based on the multiple predicted catalyst structure vectors and multiple predicted catalyst performance vectors.

[0159] In detail, the modules in the catalyst synthesis parameter optimization system 100 based on denitrification index evaluation described in this embodiment of the invention adopt the same methods as described above when in use. Figure 1 The method described herein is the same as the catalyst synthesis parameter optimization method based on denitrification index evaluation, and can produce the same technical effect, so it will not be repeated here.

[0160] like Figure 3The diagram shown is a schematic diagram of an electronic device for implementing a method for optimizing catalyst synthesis parameters based on denitrification index evaluation, according to an embodiment of the present invention.

[0161] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and capable of running on the processor 10, such as a catalyst synthesis parameter optimization method program based on denitrification index evaluation.

[0162] The memory 11 includes at least one type of readable storage medium, such as flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as a portable hard drive. In other embodiments, the memory 11 can be an external storage device of the electronic device 1, such as a plug-in portable hard drive, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device 1. Furthermore, the memory 11 includes both internal storage units and external storage devices of the electronic device 1. The memory 11 can be used not only to store application software and various types of data installed on the electronic device 1, such as the code of a catalyst synthesis parameter optimization method program based on denitrification index evaluation, but also to temporarily store data that has been output or will be output.

[0163] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., a catalyst synthesis parameter optimization method program based on denitrification index evaluation), and calls data stored in the memory 11 to perform various functions of the electronic device 1 and process data.

[0164] The bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 can be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize the connection and communication between the memory 11 and at least one processor 10, etc.

[0165] Figure 3 Only electronic devices with components are shown; it will be understood by those skilled in the art that... Figure 3 The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0166] For example, although not shown, the electronic device 1 may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management system, thereby enabling functions such as charging management, discharging management, and power consumption management through the power management system. The power supply may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0167] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), which is typically used to establish communication connections between the electronic device 1 and other electronic devices.

[0168] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), and optionally, a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device 1 and to display a visual user interface.

[0169] The catalyst synthesis parameter optimization method program based on denitrification index evaluation stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When run in the processor 10, it can achieve the following:

[0170] The catalyst product to be synthesized and the low-titanium blast furnace slag were identified, and a set of candidate metal elements was set based on the catalyst product to be synthesized and the low-titanium blast furnace slag.

[0171] The candidate metal elements are extracted sequentially from the candidate metal element set. Based on the candidate metal elements, multiple available metal raw materials are queried. Catalysts are synthesized according to the multiple available metal raw materials to obtain multiple test catalyst groups. The multiple test catalyst groups corresponding to the candidate metal elements are merged to obtain the test catalyst set.

[0172] Multiple denitrification scenarios and denitrification performance evaluation standards were identified. Based on these scenarios and standards, catalytic simulations were performed on the test catalyst set to obtain a catalyst denitrification dataset.

[0173] Deep learning was performed using a catalyst denitrification dataset to obtain a catalyst component selection model;

[0174] Based on the preset target denitrification scenario, a target scenario feature vector is constructed. The target scenario feature vector is then input into the catalyst component selection model to obtain multiple predicted catalyst structure vectors and multiple predicted catalyst performance vectors.

[0175] The target catalyst is selected based on multiple predicted catalyst structure vectors and multiple predicted catalyst performance vectors, and the catalyst synthesis parameters are optimized based on the denitrification index evaluation based on the target catalyst.

[0176] Specifically, the processor 10's implementation method for the above instructions can be found in [reference needed]. Figures 1 to 3 The descriptions of the relevant steps in the corresponding embodiments are not repeated here.

[0177] Furthermore, if the modules / units integrated in the electronic device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or system capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0178] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device, can perform the following:

[0179] The catalyst product to be synthesized and the low-titanium blast furnace slag were identified, and a set of candidate metal elements was set based on the catalyst product to be synthesized and the low-titanium blast furnace slag.

[0180] The candidate metal elements are extracted sequentially from the candidate metal element set. Based on the candidate metal elements, multiple available metal raw materials are queried. Catalysts are synthesized according to the multiple available metal raw materials to obtain multiple test catalyst groups. The multiple test catalyst groups corresponding to the candidate metal elements are merged to obtain the test catalyst set.

[0181] Multiple denitrification scenarios and denitrification performance evaluation standards were identified. Based on these scenarios and standards, catalytic simulations were performed on the test catalyst set to obtain a catalyst denitrification dataset.

[0182] Deep learning was performed using a catalyst denitrification dataset to obtain a catalyst component selection model;

[0183] Based on the preset target denitrification scenario, a target scenario feature vector is constructed. The target scenario feature vector is then input into the catalyst component selection model to obtain multiple predicted catalyst structure vectors and multiple predicted catalyst performance vectors.

[0184] The target catalyst is selected based on multiple predicted catalyst structure vectors and multiple predicted catalyst performance vectors, and the catalyst synthesis parameters are optimized based on the denitrification index evaluation based on the target catalyst.

[0185] In the embodiments provided by this invention, it should be understood that the disclosed devices, systems, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative, and actual implementations may have other classification methods.

[0186] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0187] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0188] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0189] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for optimizing catalyst synthesis parameters based on denitrification index evaluation, characterized in that, The method includes: The catalyst product to be synthesized and the low-titanium blast furnace slag were identified, and a set of candidate metal elements was set based on the catalyst product to be synthesized and the low-titanium blast furnace slag. The candidate metal elements are extracted sequentially from the candidate metal element set. Based on the candidate metal elements, multiple available metal raw materials are queried. Catalysts are synthesized according to the multiple available metal raw materials to obtain multiple test catalyst groups. The multiple test catalyst groups corresponding to the candidate metal elements are merged to obtain the test catalyst set. Multiple denitrification scenarios and denitrification performance evaluation standards were identified. Based on these scenarios and standards, catalytic simulations were performed on the test catalyst set to obtain a catalyst denitrification dataset. Deep learning was performed using a catalyst denitrification dataset to obtain a catalyst component selection model; Based on the preset target denitrification scenario, a target scenario feature vector is constructed. The target scenario feature vector is then input into the catalyst component selection model to obtain multiple predicted catalyst structure vectors and multiple predicted catalyst performance vectors. The target catalyst is selected based on multiple predicted catalyst structure vectors and multiple predicted catalyst performance vectors, and the catalyst synthesis parameters are optimized based on the denitrification index evaluation based on the target catalyst.

2. The method for optimizing catalyst synthesis parameters based on denitrification index evaluation as described in claim 1, characterized in that, The process involves synthesizing catalysts from multiple available metal raw materials to obtain multiple test catalyst groups, including: Available metal materials are extracted sequentially from multiple available metal materials, and the following operations are performed on the extracted available metal materials: The theoretical proportions of available metal raw materials are set to obtain the catalyst raw material proportion set. Catalysts were synthesized according to the catalyst raw material ratio group to obtain the test catalyst group, wherein the proportion of metal raw materials in the catalyst raw material ratio group corresponds one-to-one with the test catalyst in the test catalyst group. By summarizing the test catalyst groups corresponding to each available metal raw material from multiple available metal raw materials, multiple test catalyst groups are obtained.

3. The method for optimizing catalyst synthesis parameters based on denitrification index evaluation as described in claim 2, characterized in that, The catalyst set was subjected to catalytic simulation based on multiple denitrification scenarios and denitrification performance evaluation standards to obtain a catalyst denitrification dataset, including: Test catalysts were extracted sequentially from the test catalyst set, and the following operations were performed on the extracted test catalysts: Query the catalyst synthesis energy consumption and catalyst synthesis efficiency during the synthesis process of the tested catalyst; Structural features were extracted from the test catalyst to obtain the simulated catalyst structure vector; Based on multiple denitrification scenarios, denitrification performance evaluation standards, catalyst synthesis energy consumption, catalyst synthesis efficiency, and simulated catalyst structure vector, multi-scenario denitrification simulations were performed on the test catalysts to obtain multiple catalyst denitrification data. By merging multiple catalyst denitrification data corresponding to each test catalyst in the test catalyst set, a catalyst denitrification dataset is obtained.

4. The method for optimizing catalyst synthesis parameters based on denitrification index evaluation as described in claim 3, characterized in that, The step of extracting structural features from the test catalyst to obtain a simulated catalyst structure vector includes: Microscopic images of the test catalyst were obtained by taking microscopic photographs. The catalyst microscopic image is preprocessed to obtain an enhanced microscopic image, wherein the preprocessing includes: noise reduction and contrast enhancement; The enhanced microscopic image is segmented to obtain a binary microscopic image; The binary microscopic image is input into a pre-trained feature extractor to obtain a simulated catalyst structure vector, wherein the feature extractor is a convolutional neural network.

5. The method for optimizing catalyst synthesis parameters based on denitrification index evaluation as described in claim 4, characterized in that, Based on multiple denitrification scenarios, denitrification performance evaluation standards, catalyst synthesis energy consumption, catalyst synthesis efficiency, and simulated catalyst structure vectors, multi-scenario denitrification simulations were performed on the test catalysts to obtain multiple catalyst denitrification data, including: Extract the denitrification scenarios sequentially from multiple scenarios requiring denitrification; Based on the denitrification scenario and the tested catalyst, scenario denitrification simulation was performed to obtain the simulation scenario feature vector and the simulation denitrification performance vector. The catalyst synthesis energy consumption and catalyst synthesis efficiency are added to the simulated denitrification performance vector to obtain the simulated catalyst performance vector. By merging the simulated catalyst structure vector, the simulated catalyst performance vector, and the simulated scenario feature vector, the original denitrification data is obtained. The validity of the original denitrification data is evaluated based on the denitrification performance evaluation standard to obtain the data validity result, where the data validity result is either valid or invalid; If the data result is valid, then the original denitrification data shall be recorded as catalyst denitrification data; By summarizing the catalyst denitrification data corresponding to the denitrification scenarios, multiple catalyst denitrification data points are obtained.

6. The method for optimizing catalyst synthesis parameters based on denitrification index evaluation as described in claim 5, characterized in that, The denitrification scenario simulation based on the denitrification scenario to be denitrified and the test catalyst is used to obtain the simulated scenario feature vector and the simulated denitrification performance vector, including: Based on the scenario to be denitrified, the composition of waste incineration flue gas was simulated to obtain simulated incineration flue gas. The scenario features of the simulated incineration flue gas were constructed to obtain the simulated scenario feature vector. The test catalyst was placed in a pre-constructed reaction tube to obtain a denitrification environment, which included a gas inlet and a gas outlet. Denitrification tests were conducted using a denitrification environment and simulated incineration flue gas to obtain the effective denitrification temperature range and effective denitrification rate. Record the amount of byproducts generated and the harmfulness index of byproducts in the denitrification test; The stable temperature range of the test catalyst is obtained, and the catalyst lifetime is tested based on the stable temperature range and the preset denitrification rate threshold to obtain the catalyst lifetime. Vectors are generated based on the effective denitrification temperature range, effective denitrification rate, byproduct generation amount, byproduct harmfulness index, and catalyst lifetime to obtain the simulated denitrification performance vector.

7. The method for optimizing catalyst synthesis parameters based on denitrification index evaluation as described in claim 6, characterized in that, The denitrification test, conducted using a denitrification environment and simulated incineration flue gas, yields the effective denitrification temperature range and effective denitrification rate, including: Obtain the catalyst mass in the denitrification environment; Based on catalyst quality, denitrification environment, and simulated incineration flue gas, a set of catalyst ambient temperature and an initial denitrification rate set were obtained. Based on the catalyst ambient temperature set and the initial denitrification rate set, curve fitting was performed to obtain the denitrification rate-temperature curve; Based on the denitrification rate threshold, the effective denitrification temperature range and effective denitrification rate are identified in the denitrification rate temperature curve.

8. The method for optimizing catalyst synthesis parameters based on denitrification index evaluation as described in claim 7, characterized in that, The catalyst ambient temperature set and initial denitrification rate set obtained based on catalyst quality, denitrification environment, and simulated incineration flue gas include: Based on the preset gas volume flow rate and the preset catalyst ambient temperature, the simulated combustion flue gas is introduced into the denitrification environment, and the gas inlet and gas outlet are continuously detected using a pre-built gas analyzer to obtain the inlet nitrification concentration sequence and the outlet nitrification concentration sequence. The initial denitrification rate was calculated based on the gas volumetric flow rate, catalyst mass, inlet nitrification concentration sequence, and outlet nitrification concentration sequence. The catalyst ambient temperature is adjusted according to a preset temperature change strategy to obtain the adjusted ambient temperature. The ambient temperature is adjusted to be the catalyst ambient temperature, and the process of introducing simulated combustion flue gas into the denitrification environment according to the preset gas volume flow rate and the preset catalyst ambient temperature is repeated until a pre-constructed stop test command is received. The catalyst ambient temperature and initial denitrification rate are summarized separately to obtain the catalyst ambient temperature set and the initial denitrification rate set.

9. The method for optimizing catalyst synthesis parameters based on denitrification index evaluation as described in claim 8, characterized in that, The step of selecting a target catalyst based on multiple predicted catalyst structure vectors and multiple predicted catalyst performance vectors includes: Based on multiple predicted catalyst structure vectors, identify multiple similar catalysts in a pre-defined catalyst structure feature table; Receive a set of catalyst performance weights, wherein the number of catalyst performance weights in the set is the same as the vector dimension of the predicted catalyst performance vector; The predicted catalyst performance vectors are extracted sequentially from multiple predicted catalyst performance vectors, and a weighted calculation is performed based on the predicted catalyst performance vectors and the catalyst performance weight set to obtain the comprehensive catalytic performance value. By summing up the comprehensive catalytic performance values, multiple comprehensive catalytic performance values ​​are obtained, and the maximum catalytic performance value among the multiple comprehensive catalytic performance values ​​is identified. Identify the target catalyst with the highest catalytic performance value among multiple similar catalysts.

10. A catalyst synthesis parameter optimization system based on denitrification index evaluation, characterized in that, The system includes: The metal element setting module is used to identify the catalyst product to be synthesized and the low-titanium blast furnace slag, and to set a set of candidate metal elements based on the catalyst product to be synthesized and the low-titanium blast furnace slag. The catalyst preparation module is used to sequentially extract candidate metal elements from the candidate metal element set, query multiple available metal raw materials based on the candidate metal elements, synthesize catalysts according to the multiple available metal raw materials, obtain multiple test catalyst groups, and merge the multiple test catalyst groups corresponding to the candidate metal elements to obtain the test catalyst set. A model building module was selected to identify multiple denitrification scenarios and denitrification performance evaluation standards. Based on these scenarios and standards, catalytic simulations were performed on the test catalyst set to obtain a catalyst denitrification dataset. Deep learning was then performed on this dataset to obtain a catalyst component selection model. The target catalyst selection module is used to construct a target scenario feature vector based on a preset target denitrification scenario, input the target scenario feature vector into the catalyst component selection model, obtain multiple predicted catalyst structure vectors and multiple predicted catalyst performance vectors, and select the target catalyst based on the multiple predicted catalyst structure vectors and multiple predicted catalyst performance vectors.

Images