Method and device for analyzing wax oil fraction molecular components and electronic equipment

Abstract

The embodiment of the invention provides a wax oil fraction molecular component analysis method, a wax oil fraction molecular component analysis device and electronic equipment. The analysis method of the wax oil fraction molecular components comprises the following steps: according to the macroscopic property of a wax oil fraction to be detected, obtaining the composition of molecules to be corrected of the wax oil fraction to be detected and the digital representation of various compound molecules contained in the wax oil fraction to be detected from an established wax oil molecule database; wherein the wax oil molecule database records the corresponding relationship between the molecular composition and the macroscopic properties of multiple wax oil fraction samples, and the digital representation of the compound molecules contained in each wax oil fraction sample; and performing iterative correction on the to-be-corrected molecular composition of the to-be-detected wax oil fraction by utilizing the macroscopic property acquisition model of the wax oil fraction based on the digital representation of various compound molecules contained in the to-be-detected wax oil fraction to obtain the molecular composition of the to-be-detected wax oil fraction. According to the method, the molecular composition of the wax oil fraction can be accurately, conveniently, quickly and uniquely obtained.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, device and electronic equipment for analyzing molecular components of wax oil fractions. Background Art

[0002] At present, in the relevant technologies for the study of wax oil molecular composition, the method of inferring the molecular composition of wax oil from the macroscopic properties of wax oil is mainly used to construct a wax oil molecular composition prediction model. For example, the easily detectable macroscopic properties of wax oil (such as density, D86 distillation range, etc.) are input into the wax oil molecular composition prediction model, and the model outputs a set of predicted molecular composition information. This macroscopic property inversion method assumes that the molecular composition of wax oil conforms to a certain probability density function, and the model construction is simple and fast. However, a set of macroscopic properties of wax oil may correspond to multiple sets of molecular compositions, that is, there is a multi-solution problem, and because the molecular composition information is obtained by using model prediction, its accuracy is low.

[0003] Therefore, how to accurately and conveniently obtain the molecular composition data of wax oil is a technical problem that needs to be solved urgently. Summary of the invention

[0004] The embodiments of the present application provide a method, device and electronic device for analyzing the molecular components of wax oil fractions, which are used to accurately and conveniently obtain the molecular composition of wax oil.

[0005] One of the embodiments of the present application provides a method for analyzing the molecular components of a wax oil fraction, the method comprising: obtaining, from an established wax oil molecule database, a molecular composition to be corrected of the wax oil fraction to be measured and a digital representation of various compound molecules contained in the wax oil fraction to be measured, according to the macroscopic properties of the wax oil fraction to be measured; wherein the wax oil molecule database records the correspondence between the molecular composition and the macroscopic properties of a plurality of wax oil fraction samples, as well as the digital representation of the compound molecules contained in each wax oil fraction sample; utilizing a wax oil fraction macroscopic property acquisition model, and based on the digital representation of various compound molecules contained in the wax oil fraction to be measured, iteratively correcting the molecular composition to be corrected of the wax oil fraction to be measured, to obtain the molecular composition of the wax oil fraction to be measured.

[0006] One of the embodiments of the present application provides an analytical device for molecular components of wax oil fractions, the device comprising: a first acquisition module, used to obtain, according to the macroscopic properties of the wax oil fraction to be measured, from an established wax oil molecule database, a molecular composition to be corrected of the wax oil fraction to be measured and a digital representation of various compound molecules contained in the wax oil fraction to be measured; wherein the wax oil molecule database records the correspondence between the molecular composition and the macroscopic properties of a plurality of wax oil fraction samples, as well as the digital representation of the compound molecules contained in each wax oil fraction sample; a second acquisition module, used to utilize a wax oil fraction macroscopic property acquisition model to iteratively correct the molecular composition to be corrected of the wax oil fraction to be measured based on the digital representation of various compound molecules contained in the wax oil fraction to be measured, so as to obtain the molecular composition of the wax oil fraction to be measured.

[0007] An embodiment of the present application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the method described above when running the program.

[0008] An embodiment of the present application provides a storage medium for storing a computer-readable program. When the computer-readable program is executed, the method described above is executed.

[0009] The above technical solution provided by the embodiment of the present application has at least the following advantages compared with the prior art:

[0010] In the embodiments provided by the present application, according to the macroscopic properties of the wax oil fraction to be tested, the molecular composition to be corrected of the wax oil fraction to be tested and the digital representation of various compound molecules contained in the wax oil fraction to be tested are obtained from the established wax oil molecule database; wherein, the wax oil molecule database records the correspondence between the molecular composition of a variety of wax oil fraction samples and their macroscopic properties, as well as the digital representation of the compound molecules contained in each wax oil fraction sample; using the wax oil fraction macroscopic property acquisition model, based on the digital representation of various compound molecules contained in the wax oil fraction to be tested, the molecular composition to be corrected of the wax oil fraction to be tested is iteratively corrected to obtain the molecular composition of the wax oil fraction to be tested. The solution to the molecular composition of the wax oil fraction to be tested is unique; the molecular composition of the wax oil fraction can be accurately, conveniently and quickly obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The present application will be further described in the form of exemplary embodiments, which will be described in detail by way of the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0012] Figure 1 is a flow chart of the method for analyzing the molecular components of wax oil fraction in Example 1 of the present application;

[0013] Figure 2 This is a specific implementation flow chart of establishing a wax oil molecule database in Example 2 of the present application;

[0014] Figure 3 This is an exemplary schematic diagram of a method for obtaining a digital representation of various compound molecules contained in a wax oil fraction sample according to the molecular composition data of the wax oil fraction sample in Example 2 of the present application;

[0015] Figure 4 is an exemplary schematic diagram of a method for obtaining the macroscopic properties of a wax oil fraction sample by using a wax oil fraction macroscopic property acquisition model according to digital representations of various compound molecules contained in the wax oil fraction sample in Example 2 of the present application;

[0016] Figure 5 is an exemplary schematic diagram of a training method for a molecular property prediction sub-model in Example 2 of the present application;

[0017] Figure 6 is an exemplary schematic diagram of a device for analyzing molecular components of wax oil fractions according to some embodiments of the present application;

[0018] Figure 7 It is a schematic diagram of an exemplary structure of an electronic device according to some embodiments of the present application. DETAILED DESCRIPTION

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of the present application. For ordinary technicians in this field, the present application can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0020] It should be understood that the "system", "device", "unit" and / or "module" used herein are a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0021] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "comprises" and "includes" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0022] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, the various steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.

[0023] For ease of understanding, the technical solution of the present application is introduced below in conjunction with the accompanying drawings and embodiments.

[0024] Embodiment 1

[0025] The present application provides a method for analyzing the molecular components of wax oil fractions. Figure 1 As shown, the following steps are included:

[0026] Step S110, according to the macroscopic properties of the wax oil fraction to be measured, the molecular composition to be corrected of the wax oil fraction to be measured and the digital representation of various compound molecules contained in the wax oil fraction to be measured are obtained from the established wax oil molecule database.

[0027] The wax oil molecular database records the correspondence between the molecular composition of various wax oil fraction samples and their macroscopic properties, as well as the digital representation of the compound molecules contained in each wax oil fraction sample.

[0028] In the specific implementation process, a preset similarity algorithm can be used to determine the similarity between the macroscopic properties of the wax oil fraction to be tested and the macroscopic properties of each wax oil fraction sample in the wax oil molecule database; the molecular composition of the wax oil fraction sample corresponding to the maximum similarity is used as the molecular composition to be corrected for the wax oil fraction to be tested; the digital representation of various compound molecules contained in the wax oil fraction sample corresponding to the maximum similarity is used as the digital representation of various compound molecules contained in the wax oil fraction to be tested.

[0029] In some embodiments, the preset similarity algorithm is a nearest neighbor algorithm, and the molecular composition to be corrected of the wax oil fraction to be tested can be obtained by the following method:

[0030] A physical property topology model is established based on the macroscopic properties of various wax oil fraction samples in the wax oil molecule database using the nearest neighbor algorithm; the spatial distances between the macroscopic properties of the wax oil fraction to be tested and the macroscopic properties of various wax oil fraction samples in the physical property topology model are determined using the nearest neighbor algorithm; the molecular composition of the wax oil fraction sample corresponding to the minimum spatial distance is used as the molecular composition to be corrected for the wax oil fraction to be tested; the digital representation of various compound molecules contained in the wax oil fraction sample corresponding to the minimum spatial distance is used as the digital representation of various compound molecules contained in the wax oil fraction to be tested.

[0031] For the specific process of establishing the wax oil molecule database, please refer to the introduction in the subsequent Example 2.

[0032] Step S120, using the wax oil fraction macroscopic property acquisition model, based on the digital representation of various compound molecules contained in the wax oil fraction to be tested, iteratively calibrate the molecular composition to be calibrated of the wax oil fraction to be tested, so as to obtain the molecular composition of the wax oil fraction to be tested.

[0033] In some embodiments, a wax oil fraction macroscopic property acquisition model can be used to process the digital representations of various compound molecules contained in the wax oil fraction to be tested to obtain macroscopic properties; a physical property deviation optimization model is established based on the difference between the processed macroscopic properties and the macroscopic properties of the wax oil fraction to be tested; and the physical property deviation optimization model is used to iteratively adjust the molecular composition to be corrected of the wax oil fraction to be tested to obtain the molecular composition of the wax oil fraction to be tested.

[0034] As an example only, the digital representation of various compound molecules contained in the wax oil fraction to be tested can be input into the wax oil fraction macroscopic property acquisition model, and the wax oil fraction macroscopic property acquisition model outputs the predicted values ​​of the macroscopic properties of the wax oil fraction to be tested, such as boiling point, critical temperature, critical pressure and eccentricity factor.

[0035] In some embodiments, the molecular composition to be corrected of the wax oil fraction to be measured can be iteratively adjusted in the following manner to obtain the molecular composition of the wax oil fraction to be measured:

[0036] The molecular composition to be corrected of the wax oil fraction to be tested is taken as the optimization variable, and the physical property deviation optimization model is iteratively solved using the simulated annealing algorithm; the optimal solution of the physical property deviation optimization model is taken as the molecular composition of the wax oil fraction to be tested.

[0037] The analytical method for the molecular components of the wax oil fraction provided in Example 1 of the present application obtains the molecular composition to be corrected of the wax oil fraction to be measured and the digital representation of various compound molecules contained in the wax oil fraction to be measured from the established wax oil molecule database according to the macroscopic properties of the wax oil fraction to be measured; the molecular composition to be corrected of the wax oil fraction to be measured is iteratively corrected based on the digital representation of various compound molecules contained in the wax oil fraction to be measured by using the wax oil fraction macroscopic property acquisition model, and obtains the molecular composition of the wax oil fraction to be measured. The solution of the molecular composition of the wax oil fraction to be measured is unique; this method can accurately, conveniently and quickly obtain the molecular composition of the wax oil fraction.

[0038] Embodiment 2

[0039] There are many types of compound structures in the wax oil fraction, and the content and chemical properties of each type of compound vary greatly. No analytical technique can analyze all the compounds at the same time. A series of analytical methods are required to achieve quantitative analysis of the molecular components of the wax oil fraction. In Example 2 provided in the present application, a large number of typical wax oil fraction samples were collected, including first-line, second-line, third-line, fourth-line wax oil, delayed coking wax oil, hydrocracking tail oil, hydrocracking feedstock oil and catalytic cracking slurry.

[0040] For the collected wax oil fraction samples, perform the following operations to finally establish a wax oil molecule database, see Figure 2 As shown, it includes the following steps:

[0041] Step S210, using high-resolution mass spectrometry analysis technology to analyze and detect the wax oil fraction sample, identify various compound molecules contained in the wax oil fraction sample, and obtain the molecular composition.

[0042] In the specific implementation process, various types of compounds can be separated from the wax oil fraction sample according to the chemical properties of different types of compounds in the wax oil fraction sample by using different compound extraction techniques (for example, component separation, chemical derivatization and chemical sensitization and other technical means); for each type of compound separated from the wax oil fraction sample (for example, saturated hydrocarbon compounds, olefin compounds, aromatic hydrocarbon compounds, basic nitrides, non-basic nitrides, sulfur-containing compounds, phenols and carboxylic acid compounds, etc.), corresponding different detection methods are used for detection to obtain the molecular composition of each type of compound in the wax oil fraction sample.

[0043] For example, saturated hydrocarbons and olefins can be separated by column chromatography, and the saturated components can be analyzed by Ag+ / +ESI HRMS to obtain the olefin molecular composition information. For another example, aromatic components can be separated by column chromatography, sulfonation reaction can be carried out, and aromatic sulfonation products can be obtained after aging reaction, neutralization reaction and filtration, and analyzed by GC-FI TOF MS.

[0044] In a specific implementation process, a preset data integration method may be used to integrate the molecular compositions of various types of compounds in the wax oil fraction sample obtained above to obtain the molecular composition of the wax oil fraction sample.

[0045] As an example only, all the molecular formulas obtained above can be quantitatively expressed in the order of nitrogen element allocation, sulfur element difference allocation, oxygen element difference allocation, and hydrocarbon compound normalization.

[0046] Step S220, obtaining digital representations of various compound molecules contained in the wax oil fraction sample according to the molecular composition of the wax oil fraction sample.

[0047] The digital representation of various compound molecules contained in the wax oil fraction sample may include, but is not limited to, chemical informatics languages ​​such as structure-oriented lumping (SOL), bond charge matrix (BEM), homology sequence matrix (MTHS), structural unit-bond charge matrix hybrid framework (SU-BEM), and SMILES. Among them, the SOL representation method is mainly the reference group contribution method, which establishes a mathematical relationship between molecular structure and properties and can accurately predict the conventional properties of molecules.

[0048] In some embodiments, SOL can be used to digitally represent various compound molecules contained in the wax oil fraction sample.

[0049] For a detailed description of obtaining digital representations of various compound molecules contained in the wax oil fraction sample based on the molecular composition data of the wax oil fraction sample, see Figure 3 The relevant content in will not be repeated here.

[0050] Step S230, according to the digital representation of various compound molecules contained in the wax oil fraction sample, the macroscopic properties of the wax oil fraction sample are obtained by using the wax oil fraction macroscopic property acquisition model.

[0051] In some embodiments, various groups contained in the compound molecule can be obtained based on the SOL representation of the compound molecule, and the macroscopic properties of the wax oil fraction sample can be obtained by processing the groups using a wax oil fraction macroscopic property acquisition model.

[0052] For a detailed description of obtaining the macroscopic properties of the wax oil fraction sample by using the wax oil fraction macroscopic property acquisition model based on the digital representation of various compound molecules contained in the wax oil fraction sample, see Figure 4 The relevant content in will not be repeated here.

[0053] Step S240, obtaining a wax oil molecular database according to the molecular composition and macroscopic properties of each wax oil fraction sample and the digital representation of various compound molecules contained in each wax oil fraction sample.

[0054] In the specific implementation process, various methods can be used to save the molecular composition of the various wax oil fraction samples obtained in the above steps and the macroscopic properties of the various wax oil fraction samples, as well as the digital representation of various compound molecules contained in each wax oil fraction sample in the database. As an example only, the wax oil molecule database can be obtained by using a key-value pair method, with the number of the wax oil fraction sample as the key, and the molecular composition and macroscopic properties of the wax oil fraction sample and the digital representation of various compound molecules contained in the wax oil fraction sample as the value.

[0055] In the embodiments provided in the present application, high-resolution mass spectrometry is used to analyze and detect various compound molecules contained in the wax oil fraction sample, thereby obtaining an accurate and detailed molecular composition of the wax oil fraction sample.

[0056] Figure 3 This is an exemplary schematic diagram of a method for obtaining a digital representation of various compound molecules contained in a wax oil fraction sample according to the molecular composition of the wax oil fraction sample as shown in some embodiments of the present application. Figure 3 As shown, the method comprises the following steps:

[0057] Step S310, obtaining first molecular formulas of various compound molecules contained in the wax oil fraction sample according to the molecular composition of the wax oil fraction sample.

[0058] For any compound molecule contained in the wax oil fraction sample, perform the following operations:

[0059] Step S320: acquiring candidate molecular structures of the compound molecules according to the types of the compound molecules.

[0060] The candidate molecular structures at least include candidate basic structures and candidate structural fragments corresponding to the compound molecules.

[0061] In the specific implementation process, a basic structure database can be established in advance according to the types of various compound molecules. As an example, for thiophene compounds, the core structures are thiophene and benzothiophene, etc., and SOL structure vectors can be used to represent these core molecular structures, and then the core molecular structures and their SOL structure vectors are stored in the basic structure database.

[0062] In the specific implementation process, the structural units constituting the complex compounds, such as aromatic rings, cycloalkane rings, side chains and other structures and their SOL structural vectors can be stored in the structural fragment database.

[0063] In some embodiments, based on the type of compound molecule, a basic structure corresponding to the type of compound molecule can be arbitrarily selected from an established basic structure database as a candidate basic structure; a structural fragment corresponding to the type can be arbitrarily selected from an established structural fragment database as a candidate structural fragment; and a candidate molecular structure of the compound molecule can be obtained based on the candidate basic structure and the candidate structural fragment.

[0064] In the specific implementation process, the selection and crossover process in the genetic algorithm can be referred to, a basic structure is randomly selected from the basic structure database, and then a structural fragment is selected from the structural fragment database, and the candidate molecular structure is obtained by combining the two.

[0065] Step S330, determining the target molecular structure according to the first molecular formula of the compound molecule and the candidate molecular structures.

[0066] In the specific implementation process, the structure-oriented lumped vector corresponding to the candidate molecular structure can be determined by the following method:

[0067] Obtaining a structure-guided lumped vector corresponding to a candidate basic structure included in the candidate molecular structure, and a structure-guided lumped vector corresponding to a candidate structural fragment included in the candidate molecular structure; calculating the sum of the structure-guided lumped vector corresponding to the candidate basic structure and the structure-guided lumped vector corresponding to the candidate structural fragment to obtain a structure-guided lumped vector corresponding to the candidate molecular structure.

[0068] In a specific implementation process, the structure-oriented lumped vector corresponding to the basic structure can be obtained from the basic structure database; and the structure-oriented lumped vector corresponding to the candidate structure fragment can be obtained from the structure fragment database.

[0069] After determining the structure-guided lumped vector corresponding to the candidate molecular structure, the second molecular formula corresponding to the candidate molecular structure can be determined based on the structure-guided lumped vector; determine whether there is a difference between the first molecular formula and the second molecular formula; if there is no difference between the first molecular formula and the second molecular formula, the candidate molecular structure can be used as the target molecular structure; if there is a difference between the first molecular formula and the second molecular formula, the target molecular structure is obtained based on the candidate molecular structure and the difference.

[0070] Furthermore, if the difference between the first molecular formula and the second molecular formula is an integer number of methylene groups, a group side chain corresponding to the integer number of methylene groups can be added to the candidate molecular structure to obtain a target molecular structure.

[0071] If the difference between the first molecular formula and the second molecular formula is a non-integer number of methylene groups, step S320 is performed again to obtain a new candidate molecular structure of the compound molecule according to the type of the compound molecule.

[0072] Step S340, determining a structure-oriented lumped vector corresponding to the target molecular structure as a digital representation of the compound molecule.

[0073] If there is no difference between the first molecular formula and the second molecular formula, the structure-guided lumped vector corresponding to the candidate molecular structure may be used as the structure-guided lumped vector corresponding to the target molecular structure.

[0074] If the difference between the first molecular formula and the second molecular formula is an integer number of methylene groups, the structure-guided lumped vector corresponding to the target molecular structure can be obtained based on the structure-guided lumped vector corresponding to the candidate molecular structure and the structure-guided lumped vector corresponding to the group side chain.

[0075] In this embodiment, the wax oil fraction macroscopic property acquisition model includes a molecular property prediction sub-model and a molecular property mixing sub-model. Figure 4 This is an exemplary schematic diagram of a method for obtaining the macroscopic properties of a wax oil fraction sample by using a wax oil fraction macroscopic property acquisition model according to digital representations of various compound molecules contained in the wax oil fraction sample according to some embodiments of the present application, comprising the following steps:

[0076] Step S410, based on the digital representation of various compound molecules contained in the wax oil fraction sample, using the molecular property prediction sub-model, obtain the predicted properties of various compound molecules contained in the wax oil fraction sample.

[0077] The molecular property prediction sub-model is a trained machine learning model used to predict the macroscopic properties of molecules. In the specific implementation process, the molecular property prediction sub-model can be constructed based on the neural network model.

[0078] In a specific implementation process, the following operations may be performed for any compound molecule contained in the wax oil fraction sample to obtain predicted properties of various compound molecules contained in the wax oil fraction sample:

[0079] According to the digital representation of the compound molecule, the group information of the molecular structure of the compound molecule is determined; wherein the group information of the molecular structure includes at least the basic group and the combined group of the molecular structure; the group information of the molecular structure of the compound molecule is input into the molecular property prediction sub-model to obtain the predicted properties of the compound molecule. In the specific implementation process, for ease of use, a mapping relationship between the SOL representation of the compound molecule and the various groups contained in the compound molecule can be established, and the SOL representation of the compound molecule can be directly processed using the wax oil fraction macroscopic property acquisition model by utilizing the mapping relationship.

[0080] Basic groups include but are not limited to methyl, methylene, etc. Joint groups can be used to identify complex molecular structures and distinguish isomers in wax oil fraction samples.

[0081] For the training method of the molecular property prediction submodel, see Figure 5 The relevant content in will not be repeated here.

[0082] Step S420, according to the predicted properties of various compound molecules contained in the wax oil fraction sample, the macroscopic properties of the wax oil fraction sample are obtained by using the molecular property mixed sub-model.

[0083] In some embodiments, there is a linear relationship between the target properties of various compound molecules contained in the wax oil fraction sample, for example, the four-component content (SARA), etc., and the macroscopic properties of the wax oil fraction sample can be obtained by linear superposition of the molecular content and the molecular properties. In this case, the molecular property mixed submodel can be expressed by the formula as follows:

[0084]

[0085] Among them, P b is the target macroscopic property of the wax oil fraction sample; i is the target property of the i-th compound molecule in the wax oil fraction sample; i is the percentage of the ith compound molecule in the wax oil fraction sample; n is the total number of various compound molecules contained in the wax oil fraction sample.

[0086] In some embodiments, there is a nonlinear relationship between the target properties of various compound molecules contained in the wax oil fraction sample, for example, properties such as density, kinematic viscosity, wax content and distillation range. The compound molecules will show positive or negative effects when mixed. It is necessary to use a pre-constructed macroscopic physical property prediction model to fit this situation and obtain accurate macroscopic properties of the wax oil fraction sample.

[0087] In some embodiments, the target property is kinematic viscosity, and the macroscopic physical property prediction model is the kinematic viscosity calculation formula shown below:

[0088] θ=Aexp B / T (2)

[0089] Wherein, θ is the kinematic viscosity, T is the temperature of the wax oil fraction sample, A is the first viscosity parameter obtained by fitting according to the experimental value, which includes multiple elements, and B is the second viscosity parameter obtained by fitting according to the experimental value, which includes multiple elements. In the related art, the first viscosity parameter A and the second viscosity parameter B can be obtained through literature records, but the accuracy is low, and the kinematic viscosity of the wax oil fraction sample obtained has a large error.

[0090] In some embodiments, a relatively accurate first viscosity parameter A and second viscosity parameter B can be obtained by:

[0091] The following viscosity parameter calculation equations are established:

[0092] lnB=4.717+0.00526·T b (3)

[0093]

[0094] Among them, T bis the boiling point of the wax oil fraction sample after 50% volume distillation, which is calculated based on the molecular composition data of the wax oil fraction sample; K is the characteristic factor, which is calculated based on the molecular composition data of the wax oil fraction sample.

[0095] In some embodiments, an objective function can be established based on the difference between the first kinematic viscosity value and the second kinematic viscosity value; wherein the first kinematic viscosity value is the kinematic viscosity of the wax oil fraction sample measured experimentally, and the second kinematic viscosity value is the kinematic viscosity of the wax oil fraction sample calculated using the kinematic viscosity calculation formula; using the partial least squares method, iteratively adjust each element in the first viscosity parameter and each element in the second viscosity parameter; the first viscosity parameter and the second viscosity parameter that minimize the value of the objective function are used as the final first viscosity parameter A and the second viscosity parameter B.

[0096] In summary of the above description, the macroscopic properties of the wax oil fraction sample are obtained according to the predicted properties of various compound molecules contained in the wax oil fraction sample, which includes two cases. For target properties with linear relationships between the properties of various compound molecules, such as four-component content (SARA), the macroscopic properties of the wax oil fraction sample can be obtained by linear superposition of molecular content and molecular properties. For target properties with no linear relationship between the properties of various compound molecules, the first comprehensive property is first obtained through the first target property of the various compound molecules, and then the second comprehensive property is obtained from the first comprehensive property. As described above, the comprehensive boiling point is first obtained through the boiling points of various compound molecules, and then the comprehensive kinematic viscosity is obtained from the comprehensive boiling point.

[0097] Figure 5 is an exemplary schematic diagram of a method for training a molecular property prediction sub-model according to some embodiments of the present application. Figure 5 As shown, the method comprises the following steps:

[0098] Step S510, obtaining a training data set.

[0099] In the specific implementation process, the training data set includes multiple training samples with labels. The training data set can be obtained in the following ways.

[0100] Obtain multiple groups and the molecular properties corresponding to each group; wherein the multiple groups include basic groups and combined groups. Use a classification model (e.g., random forest, etc.) to screen out effective groups from the multiple groups; wherein the contribution of the effective groups to the target macroscopic properties is greater than a preset threshold. Use the effective groups as training samples; use the molecular properties corresponding to the effective groups as labels for the training samples. A training data set is composed of multiple labeled training samples.

[0101] Step S520: using the training data set, training the initial molecular property prediction sub-model to obtain the molecular property prediction sub-model.

[0102] In some embodiments, the current molecular property prediction submodel can be used to process training samples in a training data set to obtain predicted properties of the molecular structure, and the initial current molecular property prediction submodel is the initial molecular property prediction submodel; according to the difference between the predicted properties of the molecular structure and the labels of the training samples, the parameters of the current molecular property prediction submodel are adjusted to obtain a new molecular property prediction submodel until the convergence conditions of the model are met, and the current molecular property prediction submodel is used as the final molecular property prediction submodel.

[0103] In the specific implementation process, the effective group used as the training sample can be input into the neural network to obtain the predicted value of the macroscopic property of the molecule. The predicted value and the label of the training sample are used to establish a loss function (for example, a root mean square error function, etc.) to calculate the difference between the predicted value and the label.

[0104] An optimization algorithm (e.g., a gradient descent algorithm) is used to continuously adjust the hyperparameters of the neural network, such as the number of hidden layers and the number of hidden layer nodes, to reduce the value of the loss function until the model converges, thereby obtaining a trained molecular property prediction model.

[0105] Figure 6 It is an exemplary schematic diagram of a device for analyzing molecular components of wax oil fractions according to some embodiments of the present application.

[0106] like Figure 6 As shown, the device for quickly analyzing the molecular components of wax oil fractions includes: a first acquisition module 610 and a second acquisition module 620 .

[0107] The first acquisition module 610 is used to obtain the molecular composition to be corrected of the wax oil fraction to be tested and the digital representation of various compound molecules contained in the wax oil fraction to be tested from the established wax oil molecular database according to the macroscopic properties of the wax oil fraction to be tested; wherein the wax oil molecular database records the correspondence between the molecular composition and the macroscopic properties of multiple wax oil fraction samples, as well as the digital representation of the compound molecules contained in each wax oil fraction sample.

[0108] The second acquisition module 620 is used to use the wax oil fraction macroscopic property acquisition model to iteratively calibrate the molecular composition to be calibrated of the wax oil fraction to be measured based on the digital representation of various compound molecules contained in the wax oil fraction to be measured, so as to obtain the molecular composition of the wax oil fraction to be measured.

[0109] In some embodiments, the first acquisition module 610 obtains the molecular composition to be corrected of the wax oil fraction to be measured and the digital representation of various compound molecules contained in the wax oil fraction to be measured from the established wax oil molecule database according to the macroscopic properties of the wax oil fraction to be measured, and is used to:

[0110] Using a preset similarity algorithm, the similarities between the macroscopic properties of the wax oil fraction to be tested and the macroscopic properties of each wax oil fraction sample in the wax oil molecule database are determined respectively; the molecular composition of the wax oil fraction sample corresponding to the maximum similarity is used as the molecular composition to be corrected of the wax oil fraction to be tested; the digital representation of various compound molecules contained in the wax oil fraction sample corresponding to the maximum similarity is used as the digital representation of various compound molecules contained in the wax oil fraction to be tested.

[0111] In some embodiments, the first acquisition module 610 obtains the molecular composition to be corrected of the wax oil fraction to be measured and the digital representation of various compound molecules contained in the wax oil fraction to be measured from the established wax oil molecule database according to the macroscopic properties of the wax oil fraction to be measured, and is used to:

[0112] A physical property topology model is established based on the macroscopic properties of various wax oil fraction samples in the wax oil molecule database using a nearest neighbor algorithm; the spatial distances between the macroscopic properties of the wax oil fraction to be tested and the macroscopic properties of various wax oil fraction samples in the physical property topology model are determined using a nearest neighbor algorithm; the molecular composition of the wax oil fraction sample corresponding to the minimum spatial distance is used as the molecular composition to be corrected of the wax oil fraction to be tested; and the digital representation of various compound molecules contained in the wax oil fraction sample corresponding to the minimum spatial distance is used as the digital representation of various compound molecules contained in the wax oil fraction to be tested.

[0113] In some embodiments, the second acquisition module 620 uses the wax oil fraction macroscopic property acquisition model to iteratively calibrate the molecular composition to be corrected of the wax oil fraction to be measured based on the digital representation of various compound molecules contained in the wax oil fraction to be measured, so as to obtain the molecular composition of the wax oil fraction to be measured, which is used to:

[0114] The macroscopic property acquisition model of the wax oil fraction is used to process the digital representation of various compound molecules contained in the wax oil fraction to be measured to obtain the macroscopic properties; a physical property deviation optimization model is established according to the difference between the macroscopic properties obtained by the processing and the macroscopic properties of the wax oil fraction to be measured; and the molecular composition to be corrected of the wax oil fraction to be measured is iteratively adjusted using the physical property deviation optimization model to obtain the molecular composition of the wax oil fraction to be measured.

[0115] In some embodiments, the second acquisition module 620 uses the physical property deviation optimization model to iteratively adjust the molecular composition to be corrected of the wax oil fraction to obtain the molecular composition of the wax oil fraction to be measured, which is used to:

[0116] The molecular composition to be corrected of the wax oil fraction to be measured is used as the optimization variable, and the physical property deviation optimization model is iteratively solved using a simulated annealing algorithm; the optimal solution of the physical property deviation optimization model is used as the molecular composition of the wax oil fraction to be measured.

[0117] In some embodiments, the above device further includes a wax oil molecule database establishment module 630, which is used to pre-establish the wax oil molecule database in the following manner:

[0118] For any wax oil fraction sample, the following operations are performed: using high-resolution mass spectrometry technology to analyze and detect the wax oil fraction sample, identify various compound molecules contained in the wax oil fraction sample, and obtain molecular composition; based on the molecular composition of the wax oil fraction sample, obtain a digital representation of various compound molecules contained in the wax oil fraction sample; based on the digital representation of various compound molecules contained in the wax oil fraction sample, use the wax oil fraction macroscopic property acquisition model to obtain the macroscopic properties of the wax oil fraction sample; based on the molecular composition and macroscopic properties of each wax oil fraction sample, and the digital representation of various compound molecules contained in each wax oil fraction sample, obtain the wax oil molecular database.

[0119] In some embodiments, the wax oil molecular database establishment module 630 uses high-resolution mass spectrometry analysis technology to analyze and detect the wax oil fraction sample, identify various compound molecules contained in the wax oil fraction sample, and obtain molecular composition for:

[0120] According to the chemical properties of different types of compounds in the wax oil fraction sample, various types of compounds are separated from the wax oil fraction sample using corresponding compound extraction techniques; for each type of compound separated from the wax oil fraction sample, corresponding detection methods are used to detect each type of compound to obtain the molecular composition of each type of compound; using a preset data integration method, the molecular composition of various types of compounds in the wax oil fraction sample is integrated to obtain the molecular composition of the wax oil fraction sample.

[0121] In some embodiments, the wax oil molecular database establishment module 630 obtains digital representations of various compound molecules contained in the wax oil fraction sample according to the molecular composition of the wax oil fraction sample, and is used to:

[0122] According to the molecular composition of the wax oil fraction sample, the first molecular formula of various compound molecules contained in the wax oil fraction sample is obtained; for any compound molecule contained in the wax oil fraction sample, the following operations are performed: according to the type of the compound molecule, the candidate molecular structure of the compound molecule is obtained; wherein the candidate molecular structure at least includes a candidate basic structure and a candidate structural fragment; according to the first molecular formula of the compound molecule and the candidate molecular structure, the target molecular structure is determined; and the structure-guided lumped vector corresponding to the target molecular structure is determined as a digital representation of the compound molecule.

[0123] In some embodiments, the wax oil molecule database establishment module 630 obtains candidate molecular structures of compound molecules according to the types of compound molecules, and is used to:

[0124] According to the type of compound molecules, a basic structure corresponding to the type of compound molecules is arbitrarily selected from an established basic structure database as a candidate basic structure; a structural fragment corresponding to the type is arbitrarily selected from an established structural fragment database as a candidate structural fragment; based on the candidate basic structures and the candidate structural fragments, a candidate molecular structure of the compound molecule is obtained.

[0125] In some embodiments, the wax oil molecule database establishment module 630, the target molecular structure is determined according to the first molecular formula of the compound molecule and the candidate molecular structure, for:

[0126] Determine a structure-guided lumped vector corresponding to the candidate molecular structure; determine a second molecular formula corresponding to the candidate molecular structure according to the structure-guided lumped vector; determine whether there is a difference between the second molecular formula and the first molecular formula; if not, use the candidate molecular structure as a target molecular structure; if so, obtain the target molecular structure according to the candidate molecular structure and the difference.

[0127] In some embodiments, the wax oil molecule database establishment module 630, the structure-oriented lumped vector corresponding to the candidate molecular structure is determined to:

[0128] Obtain a structure-guided lumped vector corresponding to a candidate basic structure included in the candidate molecular structure, and a structure-guided lumped vector corresponding to a candidate structural fragment included in the candidate molecular structure; calculate the sum of the structure-guided lumped vector corresponding to the candidate basic structure and the structure-guided lumped vector corresponding to the candidate structural fragment to obtain the structure-guided lumped vector corresponding to the candidate molecular structure.

[0129] In some embodiments, the wax oil molecular database establishment module 630 obtains the target molecular structure according to the candidate molecular structure and the difference, and is used to:

[0130] If the difference is an integer number of methylene groups, a group side chain corresponding to the integer number of methylene groups is added to the candidate molecular structure to obtain the target molecular structure; if the difference is a non-integer number of methylene groups, the candidate molecular structure of the compound molecule is obtained according to the type of the compound molecule.

[0131] In some embodiments, the wax oil molecule database establishment module 630, the structure-guided lumped vector corresponding to the target molecular structure is determined to:

[0132] If there is no difference between the first molecular formula and the second molecular formula, the structure-guided lumped vector corresponding to the candidate molecular structure is used as the structure-guided lumped vector corresponding to the target molecular structure; if the difference between the first molecular formula and the second molecular formula is an integer number of methylene groups, the structure-guided lumped vector corresponding to the target molecular structure is obtained based on the structure-guided lumped vector corresponding to the candidate molecular structure and the structure-guided lumped vector corresponding to the group side chain.

[0133] In some embodiments, the wax oil fraction macroscopic property acquisition model includes a molecular property prediction submodel and a molecular property mixing submodel; the wax oil molecular database establishment module 630, according to the digital representation of various compound molecules contained in the wax oil fraction sample, uses the wax oil fraction macroscopic property acquisition model to obtain the macroscopic properties of the wax oil fraction sample, for:

[0134] According to the digital representation of various compound molecules contained in the wax oil fraction sample, using the molecular property prediction sub-model, the predicted properties of various compound molecules contained in the wax oil fraction sample are obtained;

[0135] According to the predicted properties of various compound molecules contained in the wax oil fraction sample, the macroscopic properties of the wax oil fraction sample are obtained by using the molecular property mixed sub-model.

[0136] In some embodiments, the wax oil molecular database establishment module 630 obtains predicted properties of various compound molecules contained in the wax oil fraction sample based on the digital representation of various compound molecules contained in the wax oil fraction sample and using the molecular property prediction sub-model, for:

[0137] For any compound molecule contained in the wax oil fraction sample, the following operations are performed: based on the digital representation of the compound molecule, the group information of the molecular structure of the compound molecule is determined; wherein the group information of the molecular structure includes at least the basic group and the combined group of the molecular structure; the group information of the molecular structure of the compound molecule is input into the molecular property prediction sub-model to obtain the predicted property of the compound molecule.

[0138] In some embodiments, the above apparatus further comprises a molecular property prediction sub-model building module 640, which is used to obtain the molecular property prediction sub-model by:

[0139] Acquire a training data set; use the training data set to train an initial molecular property prediction sub-model to obtain the molecular property prediction sub-model.

[0140] In some embodiments, the molecular property prediction sub-model building module 640, the training data set includes a plurality of training samples carrying labels, and the acquisition of the training data set is used to:

[0141] Acquire multiple groups and the molecular properties corresponding to each group; wherein each acquired group includes a basic group and a combined group; use a classification model to screen out effective groups from the multiple groups; wherein the contribution of the effective groups to the target properties is greater than a preset threshold; use the effective groups as training samples; use the molecular properties corresponding to the effective groups as labels of the training samples; and a training data set is formed by multiple labeled training samples.

[0142] In some embodiments, the molecular property prediction sub-model building module 640 uses the training data set to train the initial molecular property prediction sub-model to obtain the molecular property prediction sub-model for:

[0143] Using the current molecular property prediction submodel, the training samples in the training data set are processed to obtain the predicted properties of the molecular structure, and the most initial current molecular property prediction submodel is the initial molecular property prediction submodel; according to the difference between the predicted properties of the molecular structure and the labels of the training samples, the parameters of the current molecular property prediction submodel are adjusted to obtain a new molecular property prediction submodel until the convergence conditions of the model are met, and the current molecular property prediction submodel is used as the final molecular property prediction submodel.

[0144] In some embodiments, there is a linear relationship between the target properties of various compound molecules contained in the wax oil fraction sample. The wax oil molecule database establishment module 630 obtains the macroscopic properties of the wax oil fraction sample by using the molecular property mixed sub-model according to the predicted properties of various compound molecules contained in the wax oil fraction sample, and is used to:

[0145] According to the predicted properties of various compound molecules contained in the wax oil fraction sample, the target macroscopic properties of the wax oil fraction sample are obtained by using the molecular property mixed sub-model as shown below:

[0146]

[0147] Among them, Pb is the target macroscopic property of the wax oil fraction sample; i is the target property of the i-th compound molecule in the wax oil fraction sample; i is the percentage content of the i-th compound molecule in the wax oil fraction sample; n is the total number of various compound molecules contained in the wax oil fraction sample.

[0148] In some embodiments, there is a nonlinear relationship between the target properties of various compound molecules contained in the wax oil fraction sample, and the molecular property mixed sub-model is a pre-constructed macroscopic physical property prediction model. The wax oil molecule database establishment module 630 obtains the macroscopic properties of the wax oil fraction sample by using the molecular property mixed sub-model according to the predicted properties of various compound molecules contained in the wax oil fraction sample, and is used for:

[0149] The predicted properties of various compound molecules contained in the wax oil fraction sample are processed by using the pre-constructed macroscopic physical property prediction model to obtain the macroscopic properties of the wax oil fraction sample.

[0150] In some embodiments, the target property is kinematic viscosity;

[0151] The molecular property mixed sub-model is the kinematic viscosity calculation formula as shown below:

[0152] θ=Aexp B / T

[0153] Wherein, θ is the kinematic viscosity, T is the temperature of the wax oil fraction sample, A is a first viscosity parameter obtained by fitting according to the experimental value, which includes multiple elements, and B is a second viscosity parameter obtained by fitting according to the experimental value, which includes multiple elements.

[0154] In some embodiments, the first viscosity parameter A and the second viscosity parameter B are obtained by:

[0155] The following viscosity parameter calculation equations are established:

[0156] ln B=4.717+0.00526·T b

[0157]

[0158] Among them, T b is the boiling point of the wax oil fraction sample after 50% volume distillation, which is calculated based on the molecular composition data of the wax oil fraction sample; K is the characteristic factor, which is calculated based on the molecular composition data of the wax oil fraction sample;

[0159] Establishing an objective function according to the difference between the first kinematic viscosity value and the second kinematic viscosity value; wherein the first kinematic viscosity value is the kinematic viscosity of the wax oil fraction sample measured according to the experiment, and the second kinematic viscosity value is the kinematic viscosity of the wax oil fraction sample calculated using the kinematic viscosity calculation formula;

[0160] Iteratively adjusting each element in the first viscosity parameter and each element in the second viscosity parameter by using a partial least squares method;

[0161] The first viscosity parameter and the second viscosity parameter when the value of the objective function is minimized are used as the final first viscosity parameter A and the second viscosity parameter B.

[0162] In the above-mentioned embodiment of the analysis device for the molecular components of wax oil fraction, the specific processing of each module and the technical effects brought about by it can be referred to the relevant description in the corresponding method embodiment, which will not be repeated here.

[0163] Figure 7 It is a schematic diagram of an exemplary structure of an electronic device according to some embodiments of the present application.

[0164] like Figure 7 As shown, the electronic device includes: at least one processor 701, at least one communication interface 702, at least one memory 703 and at least one communication bus 704; optionally, the communication interface 702 may be an interface of a communication module, such as an interface of a GSM module; the processor 701 may be a processor CPU, or an application-specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present invention. The memory 703 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. The memory 703 stores a program, and the processor 701 calls the program stored in the memory 703 to execute part or all of the above-mentioned method embodiments.

[0165] The present application relates to a storage medium for storing a computer-readable program. When the computer-readable program is executed, part or all of the above-mentioned method embodiments are executed.

[0166] Alternatively, the storage medium may be a non-transitory computer-readable storage medium, for example, the non-transitory computer-readable storage medium may be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

[0167] Based on the same inventive concept, an embodiment of the present application further provides a computer program product, including a computer program, which implements part or all of the above-mentioned method embodiments when executed by a processor.

[0168] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of the present application. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements and amendments to the present application. Such modifications, improvements and amendments are suggested in the present application, so such modifications, improvements and amendments still belong to the spirit and scope of the exemplary embodiments of the present application.

[0169] At the same time, the present application uses specific words to describe the embodiments of the present application. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of the present application. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or multiple times in different positions in the present application does not necessarily refer to the same embodiment. In addition, some features, structures or characteristics in one or more embodiments of the present application can be appropriately combined.

[0170] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in this application, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of this application. Although the above disclosure discusses some invention embodiments that are currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of this application. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.

[0171] Similarly, it should be noted that in order to simplify the description of the disclosure of this application and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this application, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the object of this application are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0172] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining digits. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of the present application are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0173] Each patent, patent application, patent application disclosure, and other materials, such as articles, books, instructions, publications, documents, etc., cited in this application are hereby incorporated by reference in their entirety. Except for application history documents that are inconsistent with or conflicting with the content of this application, documents that limit the broadest scope of the claims of this application (currently or later attached to this application) are also excluded. It should be noted that if the descriptions, definitions, and / or use of terms in the attached materials of this application are inconsistent or conflicting with the content described in this application, the descriptions, definitions, and / or use of terms in this application shall prevail.

[0174] Finally, it should be understood that the embodiments described in this application are only used to illustrate the principles of the embodiments of the present application. Other variations may also fall within the scope of the present application. Therefore, as an example and not a limitation, the alternative configurations of the embodiments of the present application may be considered to be consistent with the teachings of the present application. Accordingly, the embodiments of the present application are not limited to the embodiments explicitly introduced and described in the present application.

Claims

1. A method for analyzing molecular components of wax oil fractions, characterized in that: The method comprises: According to the macroscopic properties of the wax oil fraction to be tested, the molecular composition to be corrected of the wax oil fraction to be tested and the digital representation of various compound molecules contained in the wax oil fraction to be tested are obtained from an established wax oil molecular database; wherein the wax oil molecular database records the correspondence between the molecular composition and the macroscopic properties of a plurality of wax oil fraction samples, as well as the digital representation of the compound molecules contained in each wax oil fraction sample; The wax oil fraction macroscopic property acquisition model is used to iteratively correct the molecular composition to be corrected of the wax oil fraction to be measured based on the digital representation of various compound molecules contained in the wax oil fraction to be measured, so as to obtain the molecular composition of the wax oil fraction to be measured.

2. The method according to claim 1, characterized in that According to the macroscopic properties of the wax oil fraction to be measured, the molecular composition to be corrected of the wax oil fraction to be measured and the digital representation of various compound molecules contained in the wax oil fraction to be measured are obtained from the established wax oil molecule database, including: Using a preset similarity algorithm, respectively determine the similarity between the macroscopic properties of the wax oil fraction to be tested and the macroscopic properties of each wax oil fraction sample in the wax oil molecule database; The molecular composition of the wax oil fraction sample corresponding to the maximum similarity is used as the molecular composition to be corrected of the wax oil fraction to be tested; The digital representation of various compound molecules contained in the wax oil fraction sample corresponding to the maximum similarity is used as the digital representation of various compound molecules contained in the wax oil fraction to be tested.

3. The method according to claim 1, characterized in that: According to the macroscopic properties of the wax oil fraction to be measured, the molecular composition to be corrected of the wax oil fraction to be measured and the digital representation of various compound molecules contained in the wax oil fraction to be measured are obtained from the established wax oil molecule database, including: Using the nearest neighbor algorithm, a physical property topology model is established based on the macroscopic properties of various wax oil fraction samples in the wax oil molecule database; Using a nearest neighbor algorithm, respectively determining the spatial distances between the macroscopic properties of the wax oil fraction to be tested and the macroscopic properties of various wax oil fraction samples in the physical property topology model; The molecular composition of the wax oil fraction sample corresponding to the minimum spatial distance is used as the molecular composition to be corrected of the wax oil fraction to be measured; The digital representation of various compound molecules contained in the wax oil fraction sample corresponding to the minimum spatial distance is used as the digital representation of various compound molecules contained in the wax oil fraction to be tested.

4. The method according to claim 1, characterized in that The method of obtaining the macroscopic properties of the wax oil fraction by using the wax oil fraction macroscopic properties acquisition model, based on the digital representation of various compound molecules contained in the wax oil fraction to be measured, iteratively corrects the molecular composition to be corrected of the wax oil fraction to be measured, and obtains the molecular composition of the wax oil fraction to be measured, including: Using a wax oil fraction macroscopic property acquisition model, digital representations of various compound molecules contained in the wax oil fraction to be measured are processed to obtain macroscopic properties; Establishing a physical property deviation optimization model according to the difference between the macroscopic properties obtained by processing and the macroscopic properties of the wax oil fraction to be tested; The physical property deviation optimization model is used to iteratively adjust the molecular composition to be corrected of the wax oil fraction to be measured, so as to obtain the molecular composition of the wax oil fraction to be measured.

5. The method according to claim 4, characterized in that The method of using the physical property deviation optimization model to iteratively adjust the molecular composition to be corrected of the wax oil fraction to be measured to obtain the molecular composition of the wax oil fraction to be measured includes: The molecular composition to be corrected of the wax oil fraction to be tested is used as an optimization variable, and a simulated annealing algorithm is used to iteratively solve the physical property deviation optimization model; The optimal solution of the physical property deviation optimization model is used as the molecular composition of the wax oil fraction to be tested.

6. The method according to claim 1, characterized in that The wax oil molecule database is pre-established in the following manner: For any wax oil fraction sample, perform the following operations: Using high-resolution mass spectrometry analysis technology to analyze and detect the wax oil fraction sample, identify various compound molecules contained in the wax oil fraction sample, and obtain the molecular composition; According to the molecular composition of the wax oil fraction sample, a digital representation of various compound molecules contained in the wax oil fraction sample is obtained; According to the digital representation of various compound molecules contained in the wax oil fraction sample, the macroscopic properties of the wax oil fraction sample are obtained by using the wax oil fraction macroscopic property acquisition model; The wax oil molecular database is obtained according to the molecular composition and macroscopic properties of each wax oil fraction sample and the digital representation of various compound molecules contained in each wax oil fraction sample.

7. The method according to claim 6, characterized in that The high-resolution mass spectrometry analysis technology is used to analyze and detect the wax oil fraction sample, identify various compound molecules contained in the wax oil fraction sample, and obtain the molecular composition, including: Separating various types of compounds from the wax oil fraction sample using corresponding compound extraction techniques according to the chemical properties of different types of compounds in the wax oil fraction sample; For each type of compound separated from the wax oil fraction sample, a corresponding detection method is used to detect each type of compound to obtain the molecular composition of each type of compound; The molecular composition of various types of compounds in the wax oil fraction sample is integrated using a preset data integration method to obtain the molecular composition of the wax oil fraction sample.

8. The method according to claim 6, characterized in that The method of obtaining digital representations of various compound molecules contained in the wax oil fraction sample according to the molecular composition of the wax oil fraction sample includes: According to the molecular composition of the wax oil fraction sample, obtaining the first molecular formula of various compound molecules contained in the wax oil fraction sample; For any compound molecule contained in the wax oil fraction sample, perform the following operations: According to the type of the compound molecule, obtaining a candidate molecular structure of the compound molecule; wherein the candidate molecular structure at least includes a candidate basic structure and a candidate structural fragment; Determining a target molecular structure according to the first molecular formula of the compound molecule and the candidate molecular structure; A structure-guided lumped vector corresponding to the target molecular structure is determined as a digital representation of the compound molecule.

9. The method according to claim 8, characterized in that The step of obtaining a candidate molecular structure of the compound molecule according to the type of the compound molecule includes: According to the type of the compound molecule, a basic structure corresponding to the type of the compound molecule is randomly selected from an established basic structure database as a candidate basic structure; From the established structural fragment database, arbitrarily select a structural fragment corresponding to the type as a candidate structural fragment; A candidate molecular structure of the compound molecule is obtained according to the candidate basic structure and the candidate structural fragment.

10. The method according to claim 9, characterized in that The step of determining the target molecular structure according to the first molecular formula of the compound molecule and the candidate molecular structure comprises: Determining a structure-guided lumped vector corresponding to the candidate molecular structure; Determining a second molecular formula corresponding to the candidate molecular structure according to the structure-guided lumped vector; determining whether there is a difference between the second molecular formula and the first molecular formula; If not, taking the candidate molecular structure as the target molecular structure; If yes, the target molecular structure is obtained according to the candidate molecular structure and the difference.

11. The method according to claim 10, characterized in that The step of determining the structure-guided lumped vector corresponding to the candidate molecular structure comprises: Obtaining a structure-oriented lumped vector corresponding to a candidate basic structure included in the candidate molecular structure, and a structure-oriented lumped vector corresponding to a candidate structural fragment included in the candidate molecular structure; The sum of the structure-guided lumped vector corresponding to the candidate basic structure and the structure-guided lumped vector corresponding to the candidate structural fragment is calculated to obtain the structure-guided lumped vector corresponding to the candidate molecular structure.

12. The method according to claim 10, characterized in that The step of obtaining the target molecular structure according to the candidate molecular structure and the difference comprises: If the difference is an integer number of methylene groups, a group side chain corresponding to the integer number of methylene groups is added to the candidate molecular structure to obtain the target molecular structure; If the difference is a non-integer number of methylene groups, the candidate molecular structure of the compound molecule is obtained according to the type of the compound molecule.

13. The method according to claim 12, characterized in that The step of determining the structure-guided lumped vector corresponding to the target molecular structure comprises: If there is no difference between the first molecular formula and the second molecular formula, taking the structure-guided lumped vector corresponding to the candidate molecular structure as the structure-guided lumped vector corresponding to the target molecular structure; If the difference between the first molecular formula and the second molecular formula is an integer number of methylene groups, the structure-guided lumped vector corresponding to the target molecular structure is obtained according to the structure-guided lumped vector corresponding to the candidate molecular structure and the structure-guided lumped vector corresponding to the group side chain.

14. The method according to claim 6, characterized in that The wax oil fraction macroscopic property acquisition model includes a molecular property prediction sub-model and a molecular property mixing sub-model; The macroscopic properties of the wax oil fraction sample are obtained by using the wax oil fraction macroscopic property acquisition model based on the digital representation of various compound molecules contained in the wax oil fraction sample, including: According to the digital representation of various compound molecules contained in the wax oil fraction sample, using the molecular property prediction sub-model, the predicted properties of various compound molecules contained in the wax oil fraction sample are obtained; According to the predicted properties of various compound molecules contained in the wax oil fraction sample, the macroscopic properties of the wax oil fraction sample are obtained by using the molecular property mixed sub-model.

15. The method according to claim 14, characterized in that The method of obtaining predicted properties of various compound molecules contained in the wax oil fraction sample by using the molecular property prediction sub-model based on the digital representation of various compound molecules contained in the wax oil fraction sample includes: For any compound molecule contained in the wax oil fraction sample, perform the following operations: Determine the group information of the molecular structure of the compound molecule according to the digital representation of the compound molecule; wherein the group information of the molecular structure at least includes the basic group and the combined group of the molecular structure; The group information of the molecular structure of the compound molecule is input into the molecular property prediction sub-model to obtain the predicted properties of the compound molecule.

16. The method according to claim 14, characterized in that The molecular property prediction submodel is obtained by the following method: Get the training dataset; The initial molecular property prediction sub-model is trained using the training data set to obtain the molecular property prediction sub-model.

17. The method according to claim 16, characterized in that The training data set includes a plurality of training samples carrying labels, and the step of obtaining the training data set includes: Obtaining multiple groups and the molecular properties corresponding to each group; wherein each obtained group includes a basic group and a combined group; Using a classification model, screening out effective groups from the multiple groups; wherein the contribution of the effective groups to the target property is greater than a preset threshold; Taking the effective groups as training samples; Using the molecular properties corresponding to the effective groups as labels of the training samples; The training data set consists of multiple labeled training samples.

18. The method according to claim 16, characterized in that The method of training the initial molecular property prediction sub-model by using the training data set to obtain the molecular property prediction sub-model comprises: Using the current molecular property prediction submodel, processing the training samples in the training data set to obtain the predicted properties of the molecular structure, the most initial current molecular property prediction submodel being the initial molecular property prediction submodel; According to the difference between the predicted properties of the molecular structure and the labels of the training samples, the parameters of the current molecular property prediction submodel are adjusted to obtain a new molecular property prediction submodel until the convergence condition of the model is met, and the current molecular property prediction submodel is used as the final molecular property prediction submodel.

19. The method according to claim 14, characterized in that There is a linear relationship between the target properties of various compound molecules contained in the wax oil fraction sample. The macroscopic properties of the wax oil fraction sample are obtained by using the molecular property mixed sub-model based on the predicted properties of various compound molecules contained in the wax oil fraction sample, including: According to the predicted properties of various compound molecules contained in the wax oil fraction sample, the target macroscopic properties of the wax oil fraction sample are obtained by using the molecular property mixed sub-model as shown below: Among them, P b is the target macroscopic property of the wax oil fraction sample; i is the target property of the i-th compound molecule in the wax oil fraction sample; i is the percentage content of the i-th compound molecule in the wax oil fraction sample; n is the total number of various compound molecules contained in the wax oil fraction sample.

20. The method according to claim 14, characterized in that There is a nonlinear relationship between the target properties of various compound molecules contained in the wax oil fraction sample. The molecular property mixed sub-model is a pre-constructed macroscopic physical property prediction model. The macroscopic properties of the wax oil fraction sample are obtained by using the molecular property mixed sub-model according to the predicted properties of various compound molecules contained in the wax oil fraction sample, including: The predicted properties of various compound molecules contained in the wax oil fraction sample are processed by using the pre-constructed macroscopic physical property prediction model to obtain the macroscopic properties of the wax oil fraction sample.

21. The method according to claim 20, characterized in that The target property is kinematic viscosity; The molecular property mixed sub-model is the kinematic viscosity calculation formula as shown below: θ=Aexp B / T Wherein, θ is the kinematic viscosity, T is the temperature of the wax oil fraction sample, A is a first viscosity parameter obtained by fitting according to the experimental value, which includes multiple elements, and B is a second viscosity parameter obtained by fitting according to the experimental value, which includes multiple elements.

22. The method according to claim 21, characterized in that The first viscosity parameter A and the second viscosity parameter B are obtained by: The following viscosity parameter calculation equations are established: lnB=4.717+0.00526 T b Among them, T b is the boiling point of the wax oil fraction sample after 50% volume distillation, which is calculated based on the molecular composition data of the wax oil fraction sample; K is the characteristic factor, which is calculated based on the molecular composition data of the wax oil fraction sample; Establishing an objective function according to the difference between the first kinematic viscosity value and the second kinematic viscosity value; wherein the first kinematic viscosity value is the kinematic viscosity of the wax oil fraction sample measured according to the experiment, and the second kinematic viscosity value is the kinematic viscosity of the wax oil fraction sample calculated using the kinematic viscosity calculation formula; Iteratively adjusting each element in the first viscosity parameter and each element in the second viscosity parameter by using a partial least squares method; The first viscosity parameter and the second viscosity parameter when the value of the objective function is minimized are used as the final first viscosity parameter A and the second viscosity parameter B.

23. A device for analyzing molecular components of wax oil fractions, characterized in that: The device comprises: The first acquisition module is used to obtain the molecular composition to be corrected of the wax oil fraction to be tested and the digital representation of various compound molecules contained in the wax oil fraction to be tested from an established wax oil molecule database according to the macroscopic properties of the wax oil fraction to be tested; wherein the wax oil molecule database records the correspondence between the molecular composition and the macroscopic properties of a plurality of wax oil fraction samples, as well as the digital representation of the compound molecules contained in each wax oil fraction sample; The second acquisition module is used to use the wax oil fraction macroscopic property acquisition model to iteratively correct the molecular composition to be corrected of the wax oil fraction to be measured based on the digital representation of various compound molecules contained in the wax oil fraction to be measured, so as to obtain the molecular composition of the wax oil fraction to be measured.

24. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor executes the method according to any one of claims 1 to 22 when running the program.

25. A storage medium for storing a computer-readable program, wherein when the computer-readable program is executed, the method according to any one of claims 1 to 22 is executed.

Images