A gasoline molecular composition prediction method and system

Abstract

The present application provides a gasoline molecular composition prediction method and system. The characteristic peak information in the infrared spectrum data and Raman spectrum data of the gasoline sample to be tested is analyzed and processed, and the interaction force and energy state simulation processing under different environmental parameters are performed to obtain the preliminary gasoline molecular combination of the gasoline sample to be tested; the physical properties of each gasoline molecule in the preliminary gasoline molecular combination under specific conditions are evaluated and processed, and the relationship between the physical properties of each gasoline molecule and the corresponding gasoline molecular structure is analyzed in combination with a pre-established prediction model to predict the target gasoline molecular combination of the gasoline sample to be tested. The present application can comprehensively and accurately predict the gasoline molecular combination of the gasoline sample to be tested.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of petrochemical technology, and in particular to a method and system for predicting the molecular composition of gasoline. Background Art

[0002] As an important fuel, gasoline plays a key role in transportation, industrial production, and energy supply. With increasingly stringent environmental regulations and growing energy demand, higher requirements are placed on the quality and performance of gasoline.

[0003] Currently, the prediction of gasoline molecular composition mainly relies on the following methods:

[0004] Chemical analysis: This method uses gas chromatography, mass spectrometry, and other instrumental analysis methods to detect the main components in gasoline. This method can provide relatively detailed chemical composition information, but it cannot accurately determine the specific arrangement of individual molecules.

[0005] Spectroscopic analysis: Using techniques such as infrared spectroscopy and Raman spectroscopy, the spectral data of gasoline samples is obtained. By analyzing the characteristic peaks, the possible molecular structure is inferred. This method is simple to operate, but the accuracy of the analysis is limited, making it difficult to distinguish similar molecular structures.

[0006] Computer simulation: This method uses molecular dynamics simulations and quantum chemical calculations to predict the physical properties of gasoline molecules under different environmental conditions. This method can provide detailed molecular dynamics behavior and physical property data, but it is computationally intensive and time-consuming.

[0007] However, although spectral analysis can provide certain molecular composition information, it cannot accurately determine the specific arrangement of individual molecules, especially in complex mixtures, making it difficult to predict the molecular composition of gasoline. Summary of the Invention

[0008] The embodiments of the present application provide a gasoline molecular composition prediction method and system to solve the problem of poor gasoline molecular composition prediction accuracy in the prior art.

[0009] In a first aspect, an embodiment of the present application provides a method for predicting gasoline molecular composition, comprising:

[0010] Obtain infrared spectrum data and Raman spectrum data of the gasoline sample to be tested;

[0011] Analyzing and processing characteristic peak information in the infrared spectrum data and the Raman spectrum data to obtain a plurality of candidate gasoline molecular structures of the gasoline sample to be tested, wherein the gasoline molecular structure refers to the specific arrangement of individual gasoline molecules;

[0012] Simulating the interaction forces and energy states of the multiple candidate gasoline molecular structures under different environmental parameters to obtain a preliminary gasoline molecular composition of the gasoline sample to be tested, wherein the gasoline molecular composition refers to a mixture composed of multiple different gasoline molecular structures;

[0013] Using molecular dynamics simulation technology, the physical properties of each gasoline molecule in the preliminary gasoline molecular composition under specific conditions are evaluated, and the relationship between the physical properties of each gasoline molecule and the corresponding gasoline molecular structure is analyzed in combination with a pre-established prediction model to predict the target gasoline molecular composition of the gasoline sample to be tested;

[0014] Based on the comparison results between the target gasoline molecular composition and the gasoline molecular composition of the standard gasoline sample, the model parameters of the prediction model are adjusted to ensure the accuracy of the predicted target gasoline molecular composition, and a prediction report of the gasoline molecular composition of the gasoline sample to be tested is generated.

[0015] In a second aspect, an embodiment of the present application provides a gasoline molecular composition prediction system, comprising:

[0016] An acquisition module, used for acquiring infrared spectrum data and Raman spectrum data of the gasoline sample to be tested;

[0017] an analysis module, configured to analyze and process characteristic peak information in the infrared spectrum data and the Raman spectrum data to obtain a plurality of candidate gasoline molecular structures of the gasoline sample to be tested, wherein the gasoline molecular structure refers to a specific arrangement of individual gasoline molecules;

[0018] a simulation module for simulating the interaction forces and energy states of the plurality of candidate gasoline molecular structures under different environmental parameters to obtain a preliminary gasoline molecular composition of the gasoline sample to be tested, wherein the gasoline molecular composition refers to a mixture composed of a plurality of different gasoline molecular structures;

[0019] a prediction module for evaluating the physical properties of each gasoline molecule in the preliminary gasoline molecular composition under specific conditions using molecular dynamics simulation technology, and analyzing the relationship between the physical properties of each gasoline molecule and the corresponding gasoline molecular structure using a pre-established prediction model to predict the target gasoline molecular composition of the gasoline sample to be tested;

[0020] An adjustment module is configured to adjust the model parameters of the prediction model based on a comparison result between the target gasoline molecular composition and the gasoline molecular composition of a standard gasoline sample to ensure the accuracy of the predicted target gasoline molecular composition and generate a prediction report for the gasoline molecular composition of the gasoline sample to be tested.

[0021] In an embodiment of the present application, infrared spectral data and Raman spectral data of a gasoline sample to be tested are obtained; characteristic peak information in the infrared spectral data and the Raman spectral data is analyzed to obtain multiple candidate gasoline molecular structures of the gasoline sample to be tested, wherein a gasoline molecular structure refers to the specific arrangement of individual gasoline molecules; interaction forces and energy states of the multiple candidate gasoline molecular structures under different environmental parameters are simulated to obtain a preliminary gasoline molecular composition of the gasoline sample to be tested, wherein a gasoline molecular composition refers to a mixture of multiple different gasoline molecular structures; molecular dynamics simulation technology is used to evaluate the physical properties of each gasoline molecule in the preliminary gasoline molecular composition under specific conditions, and the relationship between the physical properties of each gasoline molecule and the corresponding gasoline molecular structure is analyzed in combination with a pre-established prediction model to predict a target gasoline molecular composition of the gasoline sample to be tested; based on the comparison results between the target gasoline molecular composition and the gasoline molecular composition of a standard gasoline sample, model parameters of the prediction model are adjusted to ensure the accuracy of the predicted target gasoline molecular composition, and a prediction report of the gasoline molecular composition of the gasoline sample to be tested is generated. This application acquires infrared and Raman spectral data from the gasoline samples to be tested, providing rich molecular structural information and laying the foundation for subsequent analysis. By analyzing the characteristic peaks in the spectral data, it accurately identifies multiple candidate gasoline molecular structures, improving prediction accuracy. Through simulation and molecular dynamics simulation, it comprehensively assesses physical properties under different environmental parameters, ensuring the reliability and practicality of the prediction results. Based on the comparison results, model parameters are adjusted to ensure the accuracy of the prediction model and generate a reliable prediction report.

[0022] Furthermore, by setting a series of environmental parameters representing actual working conditions, the present application can simulate physical properties under different conditions, thereby improving the applicability of the prediction. By simulating the dynamic behavior of gasoline molecules at the microscopic level, a detailed physical property evaluation is provided. Key indicators are extracted from the dynamic behavior data to generate physical property evaluation results for each gasoline molecule, ensuring the accuracy and reliability of the evaluation. Through the mapping relationship model, each gasoline molecule in the preliminary gasoline molecule combination is evaluated, and gasoline molecules that exhibit the best physical properties under specific conditions are screened out to ensure the optimal performance of the target gasoline molecule combination, thereby ensuring the accuracy of the predicted target gasoline molecule combination.

[0023] These and other aspects of the present application will become more readily apparent from the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0025] Figure 1 A flowchart of a gasoline molecular composition prediction method provided in an embodiment of the present application;

[0026] Figure 2 A schematic diagram of the structure of a gasoline molecular composition prediction system provided in an embodiment of the present application;

[0027] Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0028] In order to enable people skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0029] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to being different types.

[0030] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0031] Figure 1 A flowchart of a gasoline molecular composition prediction method is provided for an embodiment of the present application. Figure 1 As shown, the method includes:

[0032] 101. Obtain infrared spectrum data and Raman spectrum data of the gasoline sample to be tested;

[0033] This step involves using appropriate instruments (such as Fourier transform infrared spectrometers (FTIR) or Raman spectrometers) to collect information about the vibrations of chemical bonds in the sample being tested. These spectra provide information about the functional groups present in the sample and their relative quantities.

[0034] In the embodiment of the present application, a small amount of gasoline to be tested is placed in a sample cell suitable for infrared or Raman detection, and then measurement is performed according to the equipment operating instructions, and the position and intensity of the infrared absorption peak and the Raman scattering signal are recorded.

[0035] 102. Analyze and process characteristic peak information in the infrared spectrum data and the Raman spectrum data to obtain a plurality of candidate gasoline molecular structures of the gasoline sample to be tested, wherein the gasoline molecular structure refers to a specific arrangement of individual gasoline molecules;

[0036] In this step, the peaks in the obtained spectral data are analyzed to identify the characteristic peaks associated with specific chemical components, and then the types of compounds that may be present are inferred.

[0037] In the examples of the present application, professional software tools (such as SpectraBase, KnowItAll, etc.) are used to input the spectrum obtained from the experiment. The software will automatically compare the standard spectra of known substances in the database to help identify the key components in the sample.

[0038] Optionally, the step 102 of "parsing and processing the characteristic peak information in the infrared spectrum data and the Raman spectrum data to obtain multiple candidate gasoline molecular structures of the gasoline sample to be tested" includes: using Fourier transform technology to perform spectral analysis on the infrared spectrum data and the Raman spectrum data to extract characteristic peak position and intensity information; based on the characteristic peak position and intensity information, using a pre-constructed spectrum database to match the gasoline molecular structure corresponding to the characteristic peak information; combining the complementary information of the infrared spectrum data and the Raman spectrum data, using multimodal data analysis technology to comprehensively evaluate the gasoline molecular structure corresponding to the characteristic peak information, and generate multiple candidate gasoline molecular structures.

[0039] Fourier transform (FT): is a mathematical transformation method used to convert time domain signals into frequency domain signals to facilitate the identification of periodic components in the signal.

[0040] Spectral database: contains the spectral characteristics of a large number of known compounds, which are used to match the spectral characteristics of unknown samples to assist in the identification of compounds.

[0041] Multimodal data analysis: Comprehensively utilize different types of data sources (in this case, infrared and Raman spectroscopy data) and analyze the relationships between them to obtain more comprehensive information.

[0042] First, Fourier transform technology is applied to the collected infrared and Raman spectral data to separate the individual frequency components from the raw signal, thereby clearly identifying the characteristic peaks in the spectrum and their corresponding intensities. This step helps to improve the accuracy of characteristic peak location during subsequent analysis. Secondly, based on the characteristic peak position and intensity information obtained in the first step, a search is performed against a preset spectral database to find the best matching known compound information. This step can quickly narrow the range of candidate molecules and lay the foundation for subsequent in-depth analysis. Finally, considering the respective strengths of infrared and Raman spectroscopy, by combining the analysis results of both, the specific compounds present in the sample can be more accurately determined. This process also involves evaluating the consistency and complementarity of the information provided by different spectroscopic techniques to ultimately generate a series of reliable candidate molecular structures.

[0043] In the examples of this application, a gasoline sample was assumed to be tested. Its infrared and Raman spectra were first obtained using FTIR and Raman spectrometers, respectively. Next, the two spectra were processed using a Fourier transform algorithm, extracting a series of significant absorption and scattering peaks. For example, a strong and broad OH stretching band was detected at approximately 3770 cm⁻¹, indicating the possible presence of alcohols; concurrently, a typical CH bending mode was detected at approximately 889 cm⁻¹, suggesting the presence of a cyclic structure. These characteristic peaks were then entered into a database containing a large number of standard spectra of hydrocarbons and other common fuel additives for comparison. After screening, several compounds with high potential were preliminarily identified, including toluene, n-heptane, and a small amount of ethanol. Finally, multimodal data analysis, combined with the independent perspectives provided by the two spectra, further confirmed the presence of the aforementioned substances in the sample. Furthermore, trace amounts of methyl tert-butyl ether (MTBE), a commonly used gasoline octane improver, were detected.

[0044] This approach, combining infrared and Raman spectroscopy with advanced data processing strategies, not only significantly improves the efficiency of identifying complex components in gasoline samples but also enhances the confidence of analytical results. This approach demonstrates significant advantages, particularly for compounds that are difficult to accurately characterize using a single spectroscopic technique, enabling more refined and comprehensive fuel quality control.

[0045] 103. Simulate the interaction forces and energy states of the multiple candidate gasoline molecular structures under different environmental parameters to obtain a preliminary gasoline molecular composition of the gasoline sample to be tested, wherein the gasoline molecular composition refers to a mixture composed of multiple different gasoline molecular structures;

[0046] In this step, based on the list of possible compounds obtained in the previous step, computational chemistry methods are used to simulate the interactions between them under various conditions, thereby screening out the main components that are most likely to constitute the actual sample.

[0047] In the examples of the present application, quantum chemistry software packages such as Gaussian are used to construct candidate molecular models, and factors such as temperature and pressure are set to run molecular mechanics or quantum mechanics simulations to observe the changing trends of intermolecular interactions under different conditions.

[0048] Optionally, the step 103 of "simulating the interaction forces and energy states of the multiple candidate gasoline molecular structures under different environmental parameters to obtain a preliminary gasoline molecular combination of the gasoline sample to be tested" includes: inputting the multiple candidate gasoline molecular structures into quantum chemical calculation software, setting a series of environmental parameters representing actual working conditions, including at least temperature and pressure; using quantum chemical calculation methods to simulate the interaction forces and energy states of each candidate gasoline molecular structure under different environmental parameters, calculating the lowest energy conformation and stability of each gasoline molecule, and using these as simulation results; based on the simulation results, screening out gasoline molecular structures with the lowest energy and stability under the set environmental parameters to form a preliminary gasoline molecular structure list; combining the preliminary gasoline molecular structure list, evaluating the relative stability of different gasoline molecular structures in the mixture and the distribution probability under different conditions; and determining the proportion of each gasoline molecule in the preliminary gasoline molecular combination based on the relative stability and the distribution probability to generate the preliminary gasoline molecular combination of the gasoline sample to be tested.

[0049] Quantum chemical calculation: a method that uses the principles of quantum mechanics to study the electronic structure and properties of atomic and molecular systems.

[0050] Lowest energy conformation: the lowest energy state in a molecule, that is, the most stable structural form.

[0051] Relative stability: the relative ability of a molecule to maintain its structure under different conditions.

[0052] Distribution probability: the probability that a certain molecule exists in a mixture under given conditions.

[0053] First, all candidate gasoline molecular structures obtained in step 102 are imported into professional quantum chemistry software such as Gaussian and ORCA. These software programs can predict various molecular properties based on quantum mechanical equations. Next, a series of environmental parameters are set for the simulation, such as temperatures ranging from -20°C to 60°C and pressures ranging from 1 atm to 5 atm. These parameters should reflect actual usage or storage conditions. Software features, such as density functional theory (DFT), are then used to calculate the energy states and interaction forces of each candidate molecule under these conditions. Based on the simulation results, molecular structures with the lowest energy and good stability under the given conditions are selected. This step helps eliminate molecular forms that are unstable or unlikely to occur under actual conditions. Statistical thermodynamic methods, such as the Boltzmann distribution, are then used to assess the relative stabilities and distribution probabilities of different molecular structures in the mixture. This step accounts for intermolecular competition and provides a better understanding of the mixture's true composition. Finally, based on the aforementioned relative stabilities and distribution probabilities, the relative abundances of each candidate molecule in the mixture under specific environmental conditions are calculated. This step ultimately yields a quantitative, preliminary gasoline molecular composition.

[0054] In the examples of this application, it is assumed that a list of candidate molecules containing several typical gasoline components (such as n-heptane, isooctane, toluene, etc.) has been obtained. The structures of these molecules are input into the Gaussian software, and two extreme working environments are set: low temperature and low pressure (-⅓ atmospheres, -20°C) and high temperature and high pressure (5 atmospheres, 60°C). Through DFT calculations, it is found that under low temperature and low pressure conditions, n-heptane presents a lower energy state and shows good stability; while under high temperature and high pressure conditions, isooctane and toluene show better stability. Based on the Boltzmann distribution formula , the probability distribution of each molecule under each condition can be calculated, where It is The energy of molecules, is the Boltzmann constant, is the absolute temperature. For example, If the energy of n-heptane , and the energy of isooctane , then the distribution probability of n-heptane at this temperature will be significantly higher than that of early octane.

[0055] This comprehensive quantum chemical calculation process not only allows for more accurate predictions of the actual forms and proportions of components in complex mixtures, but also reveals how these components are affected by external conditions. This approach provides a higher level of understanding than traditional methods, contributing to the development of a new generation of fuel products with improved performance and greater adaptability. Furthermore, it provides a theoretical basis for further optimizing production processes.

[0056] 104. Using molecular dynamics simulation technology, evaluate the physical properties of each gasoline molecule in the preliminary gasoline molecular composition under specific conditions, and analyze the relationship between the physical properties of each gasoline molecule and the corresponding gasoline molecular structure in combination with a pre-established prediction model to predict the target gasoline molecular composition of the gasoline sample to be tested;

[0057] In this step, the behavior patterns of the selected system are further studied through molecular dynamics simulations on long time scales, and the physical properties of each molecule such as density and viscosity are evaluated.

[0058] In the examples of the present application, a simulation box is established on a platform such as LAMMPS, an appropriate number of target molecules are added, and a long-term simulation is started after setting boundary conditions. During this period, changes in the macroscopic properties of the system are monitored and compared with theoretical values ​​to verify the validity of the model.

[0059] Optionally, the step 104 of "using molecular dynamics simulation technology to evaluate the physical properties of each gasoline molecule in the preliminary gasoline molecule combination under specific conditions" includes: inputting the preliminary gasoline molecule combination into molecular dynamics simulation software, and setting a series of environmental parameters representing actual working conditions, including at least temperature and pressure; under the environmental parameters, simulating the dynamic behavior of each gasoline molecule in the preliminary gasoline molecule combination at a microscopic level to calculate the physical properties of each gasoline molecule, wherein the dynamic behavior includes at least the interaction between gasoline molecules, the diffusion process, and the aggregation morphology; and extracting key indicators from the physical properties of each gasoline molecule to generate an evaluation result of the physical properties of each gasoline molecule.

[0060] Molecular dynamics simulation: A computer simulation technique that studies the behavior of matter at the atomic or molecular level by solving Newton's equations of motion to track the motion of large numbers of particles over time.

[0061] Dynamic behavior: In molecular dynamics simulations, this refers to the interactions between molecules (such as van der Waals forces, hydrogen bonds, etc.), diffusion processes (how molecules move in a medium), and aggregation morphology (how molecules aggregate to form larger structures).

[0062] Physical properties: including but not limited to density, viscosity, surface tension, etc., which can be directly or indirectly calculated through molecular dynamics simulation.

[0063] Key indicators: important parameters extracted from simulation results to characterize the physical properties of molecules.

[0064] First, a preliminary gasoline molecule composition is input into molecular dynamics software. The preliminary gasoline molecule composition obtained in step 103 is input into molecular dynamics simulation software, such as LAMMPS or GROMACS. These software programs can simulate the dynamic evolution of molecular systems over time. Next, a series of environmental parameters representing actual operating conditions are set, such as temperature (e.g., 20°C, 60°C) and pressure (e.g., 1 atm, 5 atm). A molecular dynamics simulation is then run to simulate the microscopic behavior of gasoline molecules under these conditions, including their interactions, diffusion processes, and aggregation patterns. Furthermore, based on the simulation results, the physical properties of each gasoline molecule under the given environmental parameters are calculated. This may involve using specific algorithms or formulas to calculate properties such as density and viscosity. For example, density can be calculated by dividing the mass of all molecules within a simulation cell by its volume. Finally, key metrics are extracted from the physical properties of each gasoline molecule. These metrics reflect the molecular behavior under specific conditions. For example, viscosity can be a key metric because it directly affects the fluidity and combustion properties of the fuel.

[0065] In this example, we assume a preliminary gasoline molecular composition consisting of n-heptane, isooctane, and toluene. This composition is input into LAMMPS, and two operating points are set: room temperature and atmospheric pressure (20°C, 1 atm) and high temperature and high pressure (60°C, ~5 atm).

[0066] Set simulation parameters: Temperature: 20°C (293.15K) and 60°C (333.15K)

[0067] Pressure: 1 atm (101.325 kPa) and 5 atm (506.625 kPa)

[0068] Simulation process: Run a LAMMPS simulation and track the position and velocity of each molecule. Use the radial distribution function (RDF) to analyze the distance distribution between molecules. Calculate the mean square displacement (MSD) to estimate the diffusion coefficient of the molecules.

[0069] Density: Calculated from total mass and volume.

[0070] Viscosity: calculated by Green-Kubo formula, ,in is the volume, is the Boltzmann constant, It's the temperature, is the xy component of the shear stress tensor.

[0071] Key indicator extraction: For each molecule, the density and viscosity were calculated under two conditions. It was found that as the temperature increased, the viscosity of all molecules decreased, but the viscosity of toluene changed most significantly.

[0072] The application of molecular dynamics simulation technology enables detailed information on the microscopic behavior of gasoline molecules under different environmental conditions. This not only helps to more accurately predict the target gasoline molecular composition, but also provides a deeper understanding of the impact of various molecules on the overall physical properties of gasoline. For example, toluene was found to be particularly sensitive to temperature changes, providing important insights for adjusting the formulation to suit different climate conditions. Furthermore, this simulation-based approach can help researchers explore the effects of new additives and accelerate the development of new materials.

[0073] Optionally, the step 104 of "analyzing the relationship between the physical properties of each gasoline molecule and the corresponding gasoline molecular structure in combination with a pre-established prediction model to predict a target gasoline molecule combination of the gasoline sample to be tested" includes: matching the physical properties with the gasoline molecular structure of each gasoline molecule in the preliminary gasoline molecule combination to form a multidimensional feature data set containing gasoline molecular structure and physical properties; based on the multidimensional feature data set, using a pre-established prediction model, analyzing the relationship between the physical properties of each gasoline molecule and the corresponding gasoline molecular structure to establish a mapping relationship model between physical properties and gasoline molecular structure; using the mapping relationship model, evaluating each gasoline molecule in the preliminary gasoline molecule combination to screen out gasoline molecules that exhibit the best physical properties under specific conditions; and combining the screened gasoline molecules to form a target gasoline molecule combination of the gasoline sample to be tested, thereby ensuring that the target gasoline molecule combination has the best physical properties under specific conditions.

[0074] Multidimensional feature dataset: A data set that contains molecular structure information (such as molecular formula, bond length, bond angle, etc.) and corresponding physical properties (such as density, viscosity, etc.).

[0075] Predictive model: A model built based on statistical learning or machine learning methods that is used to predict the output of unseen data, specifically a model that predicts physical properties from molecular structure.

[0076] Mapping relationship model: a mathematical or logical expression that describes the relationship between molecular structural characteristics and physical properties.

[0077] Optimal physical properties: refers to physical properties that exhibit excellent performance in specific application scenarios, such as high combustion efficiency and moderate volatility.

[0078] First, the physical properties of each gasoline molecule obtained through molecular dynamics simulation in step 103 are linked to its structural information to create a dataset containing molecular structures and physical properties. This dataset should be rich enough to cover a wide range of molecular structure types and their corresponding physical properties. Next, a machine learning algorithm (such as random forest, support vector machine, neural network, etc.) is used to train the multidimensional feature dataset to build a model that can accurately predict physical properties. The model training process includes selecting appropriate features, dividing the training and test sets, and adjusting hyperparameters. Furthermore, using the established mapping relationship model, each molecule in the preliminary gasoline molecule combination is evaluated, and the best-performing molecules are selected based on their physical properties under specific conditions. Finally, the gasoline molecules with the best physical properties are combined to form a target gasoline molecule combination. This combination is expected to exhibit the optimal physical properties under specific conditions.

[0079] In the examples of this application, it is assumed that a preliminary gasoline molecule composition has been obtained through molecular dynamics simulation, including n-heptane, isooctane, and toluene. The goal is to find a gasoline formula with optimal volatility and combustion efficiency at 20°C. First, structural information for each molecule, such as molecular formula, molecular weight, and carbon chain length, is collected. At the same time, the physical properties of each molecule at 20°C, such as boiling point, flash point, and viscosity, are also recorded. Next, a random forest is selected as the prediction model. The model is trained using the Python scikit-learn library. Feature selection is molecular weight, carbon chain length, and boiling point. The training set to test set ratio is 70%:30%. Finally, the model parameters are adjusted through cross-validation. The trained random forest model is further applied to predict the volatility and combustion efficiency of each molecule at 20°C. The prediction results show that isooctane performs better in terms of volatility, while toluene performs better in terms of combustion efficiency. Finally, isooctane and toluene are combined to form a target gasoline molecule composition with both good volatility and high combustion efficiency.

[0080] This predictive model-based approach not only enables more accurate predictions of optimal gasoline molecular combinations under specific conditions, but also provides a deeper understanding of the mechanisms by which molecular structure influences physical properties. This provides a scientific basis for the development of high-performance fuels and opens new avenues for the design of future fuels. Furthermore, this approach is highly versatile and can be applied to molecular design and performance optimization in other similar fields.

[0081] 105. Based on the comparison result between the target gasoline molecular composition and the gasoline molecular composition of the standard gasoline sample, adjust the model parameters of the prediction model to ensure the accuracy of the predicted target gasoline molecular composition, and generate a prediction report of the gasoline molecular composition of the gasoline sample to be tested.

[0082] In this step, the last step is to optimize the prediction algorithm based on the results of all the previous steps to ensure that the final output result is as close to the actual situation as possible; at the same time, the data of the entire process is organized into a formal document.

[0083] In the embodiments of the present application, the results of the comparison with the standard sample are fed back to the machine learning framework (such as support vector machine SVM, random forest RF, etc.), and the training set weight coefficient is fine-tuned until the prediction error is minimized; then a detailed project summary report is written, including but not limited to the experimental design, main findings, conclusions and suggestions, etc.

[0084] Optionally, the method of simulating the dynamic behavior of each gasoline molecule in the preliminary gasoline molecule combination at a microscopic level under the environmental parameters to calculate the physical properties of each gasoline molecule includes: initializing a molecular dynamics simulation system under the environmental parameters to assign an initial position and an initial velocity to each gasoline molecule; simulating the dynamic behavior of each gasoline molecule in the preliminary gasoline molecule combination at a microscopic level through molecular dynamics simulation software, and recording the position and velocity of each gasoline molecule as it changes with time during the simulation to obtain dynamic behavior data; and calculating the physical properties of each gasoline molecule based on the dynamic behavior data, wherein the physical properties include at least diffusion coefficient, viscosity, and density.

[0085] Molecular dynamics simulation: A computer simulation method used to study the motion laws of atoms or molecules in a material system as they evolve over time.

[0086] Diffusion coefficient: a physical quantity that measures the diffusion ability of a substance in a medium.

[0087] Viscosity: The manifestation of internal friction when a fluid flows, reflecting the fluid's ability to resist shear deformation.

[0088] Density: The mass of a substance contained in a unit volume.

[0089] First, a virtual simulation environment is created based on predefined environmental parameters such as temperature and pressure. A starting position and velocity are randomly assigned to each type of gasoline molecule. Next, a specialized MD software package is used to run the simulation, tracking the positional changes of all particles over a given timeframe and recording their velocities. Finally, based on the collected data set, an appropriate algorithm or model is applied to estimate the specific value of the target physical quantity.

[0090] In the examples of this application, it is assumed that the effect of a new gasoline additive on the overall performance of its mixture needs to be evaluated. First, 298K is set as the base temperature condition under standard atmospheric pressure, and then the MD model containing the base gasoline composition and the additive to be tested is constructed using GROMACS software. After a sufficiently long simulation time (e.g., nanosecond level), the coordinates and velocity data of all molecules in the entire time period can be extracted from the trajectory file. Then, the Einstein relationship is used To calculate the diffusion coefficient ,in express The position vector of the particle at time . Similarly, the viscosity value can be obtained using the Green-Kubo formula, and the average density of the mixture can be estimated using the particle number density.

[0091] By using this MD simulation method, not only can experimental costs and time be significantly reduced, but it can also obtain results with higher resolution than traditional experimental methods. This is of great significance for accelerating the development cycle of new products and improving product quality. In addition, this method allows researchers to explore the behavior of materials under more extreme conditions, opening up new possibilities for future technological innovation.

[0092] This application considers that calculating the physical properties of gasoline molecules (such as diffusion coefficient, viscosity, and density) in molecular dynamics simulations is an important means of evaluating their behavior under specific conditions. These physical properties directly affect gasoline's combustion efficiency, transmission performance, and environmental impact. While traditional experimental methods can provide accurate data, they are time-consuming, costly, and difficult to conduct large-scale testing under diverse conditions. Therefore, calculating these physical properties through molecular dynamics simulations has become an efficient method.

[0093] In order to ensure the accuracy and reliability of the calculation results, this method adopts the following design principles:

[0094] Temperature and pressure correction: Considering the influence of temperature and pressure on physical properties, temperature and pressure correction factors are introduced to ensure the consistency and accuracy of calculation results under different environmental conditions.

[0095] Multi-time point sampling: By sampling at multiple time points, the statistical reliability of the calculation results is ensured.

[0096] The options are as follows:

[0097] Optionally, the calculating of the physical properties of each gasoline molecule based on the dynamic behavior data includes:

[0098] The diffusion coefficient of each gasoline molecule is calculated using the following formula:

[0099] ;

[0100] in, represents the mean square displacement, which is expressed in time The average value of the square distance of the gasoline molecule's position relative to the initial position at time , represents the mean square displacement after temperature correction, represents the mean square displacement after temperature and pressure correction; Indicates the current temperature; Indicates the reference temperature; represents activation energy; Represents the gas constant, which is ; Indicates current pressure; Indicates reference pressure; Indicates the number of dimensions. For a three-dimensional system, Indicates time; represents the diffusion coefficient;

[0101] The reasons for designing the above formula include: the mean square displacement represents the square of the average distance that particles move over time. The mean square displacement is an important indicator for measuring the degree of molecular diffusion. The average value of the square distance of the position of the molecule relative to the initial position when the temperature is 0.05. Taking activation energy into consideration Pressure correction: Pressure also affects the diffusion coefficient, so a pressure correction factor is introduced. To take into account the effect of pressure. Diffusion coefficient: The final diffusion coefficient The corrected mean square displacement and time Calculated.

[0102] in, The position of each gasoline molecule changing with time is recorded by molecular dynamics simulation software, and the mean square displacement is calculated. : Current temperature, set by experiment or simulation. : Reference temperature, usually the temperature under standard conditions. : Activation energy, obtained through experiments or literature data. : Gas constant, value is P: Current pressure, set by experiment or simulation. : Reference pressure, usually the pressure under standard conditions. Indicates the number of dimensions. For a three-dimensional system, . : time, the time point recorded by the simulation.

[0103] The viscosity of each gasoline molecule is calculated using the following formula:

[0104] ;

[0105] in, represents the temperature-corrected shear stress autocorrelation function, represents the shear stress autocorrelation function, which indicates the time The average value of the product of the shear stress tensor at time t and the shear stress tensor at the initial moment, represents the shear stress autocorrelation function after temperature and pressure correction; Indicates the current temperature; Indicates the reference temperature; represents the viscosity activation energy; Represents the gas constant, which is ; Indicates current pressure; Indicates reference pressure; represents the pressure activation energy; Represents the volume of the simulation box; represents the Boltzmann constant, whose value is ; Indicates the current temperature; Indicates viscosity;

[0106] The reasons for designing this formula include: Shear stress autocorrelation function: reflects the relationship between shear stress and time. Shear stress autocorrelation function is an important indicator for measuring fluid viscosity, which indicates the relationship between shear stress and time. The average value of the product of the shear stress tensor at the time and the shear stress tensor at the initial time. Temperature correction: Temperature has a significant effect on viscosity, so a temperature correction factor is introduced. Considering the viscosity activation energy Pressure correction: Pressure also affects viscosity, so a pressure correction factor is introduced. To take into account the effect of pressure. Viscosity: Final viscosity The corrected shear stress autocorrelation function and time integral.

[0107] in, : The shear stress tensor of each gasoline molecule that changes with time is recorded by molecular dynamics simulation software, and the shear stress autocorrelation function is calculated. : Current temperature, set by experiment or simulation. : Reference temperature, usually the temperature under standard conditions. : Viscosity activation energy, obtained through experiments or literature data. : Gas constant, value is . : Current pressure, set by experiment or simulation. : Reference pressure, usually the pressure under standard conditions. : Pressure activation energy, obtained through experiments or literature data. : The volume of the simulated box, set by simulation. : Boltzmann constant, value is .

[0108] The density of each gasoline molecule is calculated using the following formula:

[0109] ;

[0110] ;

[0111] in, Indicates time When the density is corrected for temperature and pressure, Indicates the A point in time, Indicates time When simulating the total mass of all gasoline molecules in the box; Represents the volume of the simulation box; Indicates the current temperature; Indicates the reference temperature; represents the linear expansion coefficient; Indicates current pressure; Indicates reference pressure; represents density, which is obtained by averaging the density over all time periods; Indicates the total number of sampling time points.

[0112] The reasons for designing this formula include: Density: Density is an important indicator to measure the mass of a substance per unit volume. Temperature correction: Temperature has a significant effect on density, so a temperature correction factor is introduced. To take into account the linear expansion coefficient Pressure correction: Pressure also affects density, so a pressure correction factor is introduced. To take into account the influence of pressure. Multi-time point sampling: By sampling at multiple time points, the statistical reliability of the calculation results is ensured.

[0113] in, :In time The total mass of all gasoline molecules in the simulation box is recorded by molecular dynamics simulation software. : The volume of the simulated box, set by simulation. : Current temperature, set by experiment or simulation. : Reference temperature, usually the temperature under standard conditions. : Linear expansion coefficient refers to the rate of increase in the relative length of a substance due to an increase in temperature, which is obtained through experiments or literature data. : Current pressure, set by experiment or simulation. : Reference pressure, usually the pressure under standard conditions. : No. time points, recorded through simulation. : The total number of sampling time points, set by simulation.

[0114] In the examples of this application, it is assumed that the effect of a new gasoline additive on a base gasoline mixture is being studied. , the activation energy of the known base gasoline , viscosity activation energy , and the linear expansion coefficient In the MD simulation, set . Assume that after simulation, for a certain moment , got .

[0115] Diffusion coefficient Calculation example:

[0116] , , .

[0117] Viscosity The calculation examples skip the complex integration part, but it can be completed automatically by the software.

[0118] density Calculation example:

[0119] .

[0120] This MD simulation-based approach, combined with specific physical property calculation formulas, provides an economical and efficient way to predict gasoline performance indicators under different operating conditions. This not only helps reduce the number of experiments and associated costs, but also enables a more detailed understanding of complex mixture systems, thereby promoting the rapid development and market launch of new products. Furthermore, this method can help researchers better understand the relationship between microstructure and macroscopic properties, promoting theoretical innovation and development in related fields.

[0121] Optionally, the method uses the mapping relationship model to evaluate each gasoline molecule in the preliminary gasoline molecule combination and screen out gasoline molecules that exhibit the best physical properties under specific conditions, including: inputting the multidimensional feature data set into the mapping relationship model to calculate a comprehensive performance score for each gasoline molecule under specific conditions, wherein the comprehensive performance score is calculated based on the weights and importance of multiple physical properties using a predefined scoring function, and the scoring function is intended to reflect the overall performance of the gasoline molecule under specific conditions; sorting all gasoline molecules in the preliminary gasoline molecule combination according to the comprehensive performance score, and screening out the top N gasoline molecules with the highest scores as the gasoline molecules that exhibit the best physical properties under specific conditions, wherein N is a preset positive integer representing the number of molecules included in the target gasoline molecule combination; the method also includes: performing a stability analysis on the top N screened gasoline molecules to exclude gasoline molecules that are unstable or have significant defects under specific conditions.

[0122] Multidimensional feature data set: a data set containing various physical and chemical properties of gasoline molecules, such as diffusion coefficient, viscosity, density, etc.

[0123] Mapping relational model: A mathematical model that converts input multidimensional feature data into output performance scores.

[0124] Overall Performance Score: A weighted score that combines multiple physical properties to assess the overall performance of gasoline molecules under specific conditions.

[0125] Scoring function: A mathematical expression used to calculate the comprehensive performance score, which can be customized according to different needs.

[0126] Stability analysis: Evaluate whether gasoline molecules are stable under specific conditions and whether there is a risk of failure under long-term use or extreme conditions.

[0127] First, a multidimensional feature dataset of gasoline molecules obtained from molecular dynamics simulations or other sources is input into a trained mapping model. Then, a scoring function is used to calculate the overall performance score of each gasoline molecule, combining the weights and importance of each physical property. The scoring function can be a linear combination or other form, for example: ,in Represents the comprehensive performance score, represent the diffusion coefficient, viscosity and density respectively, is the corresponding weight coefficient. All gasoline molecules are further ranked according to their comprehensive performance scores, and the top N molecules with the highest scores are selected as candidate optimal molecular combinations. Finally, these top N molecules undergo further stability testing to ensure they will not decompose or otherwise react adversely in their intended operating environment.

[0128] In this example, it is assumed that the most suitable molecule for low-temperature starting performance is to be found among a series of new gasoline additives. The diffusion coefficient of each additive has been obtained through MD simulation. , viscosity and density Data. The weight coefficients are set as , to emphasize the importance of the diffusion coefficient.

[0129] Assume that the data of an additive is: .

[0130] Calculate its comprehensive performance score as:

[0131] .

[0132] After repeating this process for all additives, you'll get a list ranked in descending order by overall performance score. If you want to select the top three best-performing additives, then select the top three from this list. Next, you'll need to test these three additives for low-temperature stability to ensure they don't crystallize or experience other issues in low-temperature environments.

[0133] This mapping-based screening approach not only enables the rapid and efficient identification of gasoline molecule combinations with optimal overall performance, but also ensures their stability and reliability under specific operating conditions. This approach significantly reduces the time and resources required for experimental screening, improving the efficiency of new product development. It also provides a scientific basis for subsequent product optimization, helping companies respond more quickly to market demand and launch higher-quality products.

[0134] This application considers the need to comprehensively consider multiple physical properties, such as diffusion coefficient, viscosity, and density, when evaluating the comprehensive performance of gasoline molecules under specific conditions. These physical properties have a significant impact on gasoline's combustion efficiency, transmission performance, and environmental impact. Traditional evaluation methods are typically based on a single physical property and cannot fully reflect the comprehensive performance of gasoline molecules. Therefore, a comprehensive evaluation method is needed that can comprehensively consider multiple physical properties and provide an overall performance score.

[0135] To ensure the accuracy and reliability of the comprehensive performance score, this method adopts the following design principles:

[0136] Comprehensive evaluation of multiple physical properties: Comprehensively consider the three physical properties of diffusion coefficient, viscosity and density to ensure the comprehensiveness of the evaluation.

[0137] Dynamic weight adjustment: By dynamically adjusting the weight of each physical property, the scoring is ensured to be flexible and adaptable.

[0138] Nonlinear transformation: Normalize the physical property values ​​through nonlinear transformation to ensure the comparability between different physical properties.

[0139] Standardized scoring: Generates a standardized comprehensive performance score through weighted summation to ensure the interpretability and comparability of the score.

[0140] The options are as follows:

[0141] Optionally, inputting the multidimensional feature dataset into the mapping relationship model to calculate the comprehensive performance score of each gasoline molecule under specific conditions includes:

[0142] Multidimensional feature dataset Input the mapping relationship model and determine the physical properties Corresponding weight ,in, represents the diffusion coefficient, Indicates viscosity, Indicates density;

[0143] Diffusion coefficient weight :

[0144] ;

[0145] in, is a parameter that controls the rate of change of the weight; the expected value of the diffusion coefficient ,in, is the Boltzmann constant ; is the current temperature (K); is the viscosity of the reference solvent (Pa-s); is the effective radius of the gasoline molecule (m); and is a constant related to the shape of gasoline molecules; is a constant related to temperature; is the activation energy ; is the gas constant ;

[0146] Viscosity weight :

[0147] ;

[0148] in, is a parameter that controls the rate of change of the weight; the expected value of viscosity ,in, is the viscosity of the reference solvent ; is a constant related to the interaction force between gasoline molecules; is the current temperature (K); is a constant related to temperature; is a constant related to pressure; The current pressure ;

[0149] Density weight :

[0150] ;

[0151] in, is a parameter that controls the rate of change of weights; the expected value of density ,in, is the molar mass of the gasoline molecule ; is the reference volume ; is a constant related to temperature; is the current temperature (K); is a constant related to pressure; The current pressure ; is a constant related to temperature;

[0152] For each physical property Perform nonlinear transformation to obtain :

[0153] ;

[0154] in, is a specific physical property value, include Any one of . It is a physical property The minimum value among all samples. It is a physical property The maximum value among all samples. Yes Take the natural logarithm.

[0155] The comprehensive performance score of each gasoline molecule under specific conditions is calculated using the following formula: :

[0156] ;

[0157] in, Physical properties The weight of the weight is usually in the range of , and usually To ensure standardization of scoring. They are the physical properties after nonlinear transformation These transformed values ​​are obtained by the previous formula Calculated.

[0158] In the above process, the Sigmoid function is used Convert physical property values ​​into weights, ensuring that the weight values ​​are in Within the range. Parameters : Parameters that control the rate of change of weights, which can be adjusted according to actual needs. Expected value : The expected value of each physical property is used to centralize the physical property values ​​and ensure the rationality of the weights.

[0159] in, : Parameters that control the rate of change of weights, determined through experiments or empirical data. : are the actual values ​​of diffusion coefficient, viscosity and density, respectively, which are calculated by molecular dynamics simulation. : are the expected values ​​of diffusion coefficient, viscosity and density, respectively, which are calculated using the above expected value calculation formulas. Each expected value formula is based on a model with clear physical meaning to ensure the reliability of the calculation results, and considers the effects of temperature and pressure on physical properties, and introduces corresponding correction factors.

[0160] In the nonlinear transformation formula, the physical property values ​​are converted to the range of [0,1] by normalization to ensure the comparability between different physical properties, and the changes in small values ​​are enhanced by taking the natural logarithm to make the score more sensitive.

[0161] In this embodiment, it is assumed that there is a series of gasoline molecule data, and it is necessary to select the molecule that is most suitable for low-temperature starting performance. (Right now ),pressure (i.e. 1 standard atmosphere). The viscosity of the solvent is known. , the molar mass of a gasoline molecule , reference volume .

[0162] Calculate the expected value of the diffusion coefficient :Assumption , .

[0163] .

[0164] Calculate the expected viscosity :Assumption ,

[0165] .

[0166] .

[0167] Calculate the expected value of density :

[0168] Assumptions .

[0169] .

[0170] Suppose there are three gasoline molecules in a sample, and their diffusion coefficient, density, and density are as follows:

[0171] Sample 1: ;

[0172] Sample 2: ;

[0173] Sample 3: ;

[0174] Assume that the parameter that controls the rate of change of weights The weight and comprehensive performance score of each sample can be calculated according to the above formula.

[0175] Weights are calculated for sample 1: , , ;

[0176] Sample 2: , , ;

[0177] Sample 3: , , ;

[0178] Nonlinear transformation for sample 1: , , .

[0179] Samples 2 and 3 are calculated similarly.

[0180] The overall performance score is: .

[0181] This approach allows the weights of various physical properties to be dynamically adjusted based on expected values ​​under specific conditions, and the scoring function to be optimized through nonlinear transformations. This approach has the advantage of more accurately reflecting the comprehensive performance of gasoline molecules under specific conditions, thereby helping to screen for the optimal molecular combination among numerous candidate molecules. This approach not only improves screening efficiency but also enhances the scientific nature and accuracy of the screening results, helping to accelerate the development and optimization of new products. Furthermore, this approach can better understand and predict the performance of gasoline molecules in different application scenarios, providing strong support for future research and development.

[0182] Figure 2 A structural diagram of a gasoline molecular composition prediction system is provided for an embodiment of the present application. Figure 2 As shown, the device includes:

[0183] An acquisition module 21 is used to acquire infrared spectrum data and Raman spectrum data of a gasoline sample to be tested;

[0184] an analysis module 22 for analyzing characteristic peak information in the infrared spectrum data and the Raman spectrum data to obtain a plurality of candidate gasoline molecular structures of the gasoline sample to be tested, wherein the gasoline molecular structure refers to the specific arrangement of individual gasoline molecules;

[0185] a simulation module 23 for simulating the interaction forces and energy states of the plurality of candidate gasoline molecular structures under different environmental parameters to obtain a preliminary gasoline molecular composition of the gasoline sample to be tested, wherein the gasoline molecular composition refers to a mixture of a plurality of different gasoline molecular structures;

[0186] Prediction module 24 is configured to evaluate the physical properties of each gasoline molecule in the preliminary gasoline molecular composition under specific conditions using molecular dynamics simulation technology, analyze the relationship between the physical properties of each gasoline molecule and the corresponding gasoline molecular structure using a pre-established prediction model, and predict the target gasoline molecular composition of the gasoline sample to be tested;

[0187] The adjustment module 25 is used to adjust the model parameters of the prediction model based on the comparison results between the target gasoline molecular composition and the gasoline molecular composition of the standard gasoline sample to ensure the accuracy of the predicted target gasoline molecular composition and generate a prediction report of the gasoline molecular composition of the gasoline sample to be tested.

[0188] Figure 2 The gasoline molecular composition prediction system can be executed Figure 1The implementation principles and technical effects of the gasoline molecular composition prediction method described in the illustrated embodiment are not further elaborated. The specific manner in which the various modules and units of the gasoline molecular composition prediction system in the aforementioned embodiment perform their operations has been described in detail in the relevant embodiments of the method and will not be further elaborated here.

[0189] In one possible design, Figure 2 A gasoline molecular composition prediction system of the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0190] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0191] The processing component 32 is used to: analyze and process characteristic peak information in the acquired infrared spectrum data and Raman spectrum data of the gasoline sample to be tested, and simulate the interaction force and energy state under different environmental parameters to obtain a preliminary gasoline molecular combination of the gasoline sample to be tested; evaluate the physical properties of each gasoline molecule in the preliminary gasoline molecular combination under specific conditions, and analyze the relationship between the physical properties of each gasoline molecule and the corresponding gasoline molecular structure in combination with a pre-established prediction model to predict the target gasoline molecular combination of the gasoline sample to be tested.

[0192] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above method.

[0193] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0194] Of course, a computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0195] The input / output interface provides an interface between the processing component and the peripheral interface module, which can be an output device, an input device, etc.

[0196] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0197] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0198] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 A gasoline molecular composition prediction method according to the illustrated embodiment.

[0199] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0200] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0201] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0202] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for predicting gasoline molecular composition, characterized in that: include: Obtain infrared spectrum data and Raman spectrum data of the gasoline sample to be tested; Analyzing and processing characteristic peak information in the infrared spectrum data and the Raman spectrum data to obtain a plurality of candidate gasoline molecular structures of the gasoline sample to be tested, wherein the gasoline molecular structure refers to a specific arrangement of a single gasoline molecule; Performing interaction force and energy state simulation processing on the multiple candidate gasoline molecular structures under different environmental parameters to obtain a preliminary gasoline molecular combination of the gasoline sample to be tested, wherein the gasoline molecular combination refers to a mixture composed of multiple different gasoline molecular structures; Using molecular dynamics simulation technology, the physical properties of each gasoline molecule in the preliminary gasoline molecular combination under specific conditions are evaluated and processed, and the relationship between the physical properties of each gasoline molecule and the corresponding gasoline molecular structure is analyzed in combination with a pre-established prediction model to predict the target gasoline molecular combination of the gasoline sample to be tested; Based on the comparison result between the target gasoline molecular combination and the gasoline molecular combination of the standard gasoline sample, the model parameters of the prediction model are adjusted to ensure the accuracy of the predicted target gasoline molecular combination, and a prediction report of the gasoline molecular combination of the gasoline sample to be tested is generated; The molecular dynamics simulation technology is used to evaluate the physical properties of each gasoline molecule in the preliminary gasoline molecule combination under specific conditions, including: Inputting the preliminary gasoline molecular combination into molecular dynamics simulation software, and setting a series of environmental parameters representing actual working conditions, including at least temperature and pressure; Under the environmental parameters, simulating the dynamic behavior of each gasoline molecule in the preliminary gasoline molecule combination at a microscopic level to calculate the physical properties of each gasoline molecule, wherein the dynamic behavior at least includes the interaction, diffusion process, and aggregation morphology between gasoline molecules; Extracting key indicators from the physical properties of each gasoline molecule to generate an evaluation result of the physical properties of each gasoline molecule; The method of simulating the dynamic behavior of each gasoline molecule in the preliminary gasoline molecule combination at a microscopic level under the environmental parameters to calculate the physical properties of each gasoline molecule includes: Under the environmental parameters, the molecular dynamics simulation system is initialized to assign an initial position and an initial velocity to each gasoline molecule; Using molecular dynamics simulation software, simulating the dynamic behavior of each gasoline molecule in the preliminary gasoline molecule combination at the microscopic level, and in the simulation process, recording the position and speed of each gasoline molecule over time to obtain dynamic behavior data; Based on the dynamic behavior data, calculating the physical properties of each gasoline molecule, the physical properties at least including diffusion coefficient, viscosity, and density; Analyzing the relationship between the physical properties of each gasoline molecule and the corresponding gasoline molecular structure in combination with a pre-established prediction model, predicting the target gasoline molecule combination of the gasoline sample to be tested, including: Matching the physical property with the gasoline molecular structure of each gasoline molecule in the preliminary gasoline molecular combination to form a multidimensional feature data set including gasoline molecular structure and physical properties; Based on the multidimensional feature data set, using a pre-established prediction model, the relationship between the physical properties of each gasoline molecule and the corresponding gasoline molecular structure is analyzed to establish a mapping relationship model between the physical properties and the gasoline molecular structure; By using the mapping relationship model, each gasoline molecule in the preliminary gasoline molecule combination is evaluated to screen out gasoline molecules that exhibit the best physical properties under specific conditions; The screened gasoline molecules are combined to form a target gasoline molecule combination of the gasoline sample to be tested, ensuring that the target gasoline molecule combination has optimal physical properties under specific conditions.

2. The method according to claim 1, characterized in that Based on the dynamic behavior data, the physical properties of each gasoline molecule are calculated, including: The diffusion coefficient of each gasoline molecule is calculated using the following formula: ； in, represents the mean square displacement, which is expressed at time The average value of the square distance of the gasoline molecule's position relative to its initial position at time , represents the mean square displacement after temperature correction, represents the mean square displacement after temperature and pressure correction; Indicates the current temperature; Indicates the reference temperature; represents the activation energy; represents the gas constant; Indicates current pressure; Indicates the reference pressure; Indicates the number of dimensions; Indicates time; represents the diffusion coefficient; The viscosity of each gasoline molecule is calculated using the following formula: ； in, represents the temperature-corrected shear stress autocorrelation function, represents the shear stress autocorrelation function, which represents the time The average value of the product of the shear stress tensor at time and the shear stress tensor at the initial time, represents the shear stress autocorrelation function after temperature and pressure correction; Indicates the current temperature; Indicates the reference temperature; represents the viscosity activation energy; represents the volume constant; Indicates current pressure; Indicates the reference pressure; represents the pressure activation energy; Represents the volume of the simulation box; represents the Boltzmann constant; Indicates the current temperature; Indicates viscosity; The density of each gasoline molecule is calculated using the following formula: ； ； in, Indicates at time When the temperature and pressure are corrected, Indicates A point in time, Indicates at time simulating the total mass of all gasoline molecules in the box; Represents the volume of the simulation box; Indicates the current temperature; Indicates the reference temperature; represents the linear expansion coefficient; Indicates current pressure; Indicates the reference pressure; represents density, which is obtained by averaging the density over all time periods; Indicates the total number of sampling time points.

3. The method according to claim 1, characterized in that The mapping relationship model is used to evaluate each gasoline molecule in the preliminary gasoline molecule combination, and the gasoline molecules that exhibit the best physical properties under specific conditions are screened out, including: Inputting the multidimensional feature data set into the mapping relationship model to calculate a comprehensive performance score of each gasoline molecule under specific conditions, wherein the comprehensive performance score is calculated based on the weights and importance of multiple physical properties through a predefined scoring function, and the scoring function is intended to reflect the overall performance of the gasoline molecule under specific conditions; According to the comprehensive performance score, all gasoline molecules in the preliminary gasoline molecule combination are sorted, and the top N gasoline molecules with the highest scores are screened out as gasoline molecules that exhibit the best physical properties under specific conditions, where N is a preset positive integer representing the number of molecules included in the target gasoline molecule combination; The method further comprises: The stability analysis is performed on the first N gasoline molecules screened to exclude gasoline molecules that are unstable under specific conditions or have significant defects.

4. The method according to claim 3, characterized in that The step of inputting the multidimensional feature data set into the mapping relationship model to calculate the comprehensive performance score of each gasoline molecule under specific conditions includes: Multidimensional feature dataset Input into the mapping relationship model and determine the physical properties respectively The corresponding weight ,in, represents the diffusion coefficient, Indicates viscosity, Indicates density; Diffusion coefficient weight Calculated by the following formula: ； in, is a parameter that controls the rate of change of the weight; the expected value of the diffusion coefficient ,in, is the Boltzmann constant; is the current temperature; is the viscosity of the reference solvent; is the effective radius of the gasoline molecule; and is a constant related to the shape of gasoline molecules; is a constant related to temperature; is the activation energy; is the gas constant; Viscosity Weight Calculated by the following formula: ； in, is the parameter that controls the rate of change of the weight; the expected value of viscosity ,in, is the viscosity of the reference solvent; is a constant related to the interaction force between gasoline molecules; is the current temperature; is a constant related to temperature; is a constant related to pressure; It is the current pressure; Density weight Calculated by the following formula: ； in, is a parameter that controls the rate of change of the weight; the expected value of the density ,in, is the molar mass of the gasoline molecule; is the reference volume; is a constant related to temperature; is the current temperature; is a constant related to pressure; It is the current pressure; is a constant related to temperature; For each physical property Perform nonlinear transformation to obtain : ； in, is a specific physical property value, include Any of the following: It is a physical property The minimum value among all samples; It is a physical property The maximum value among all samples; Yes Take the natural logarithm; The comprehensive performance score of each gasoline molecule under specific conditions is calculated by the following formula : ； in, The physical properties are The weight of the weight is usually , and usually To ensure standardization of scoring; They are the physical properties after nonlinear transformation. The value of .

5. The method according to claim 1, characterized in that The step of analyzing and processing the characteristic peak information in the infrared spectrum data and the Raman spectrum data to obtain a plurality of candidate gasoline molecular structures of the gasoline sample to be tested includes: Using Fourier transform technology, the infrared spectrum data and the Raman spectrum data are subjected to spectrum analysis to extract characteristic peak position and intensity information; Based on the characteristic peak position and intensity information, using a pre-built spectrum database, matching the gasoline molecular structure corresponding to the characteristic peak information; The complementary information of the infrared spectrum data and the Raman spectrum data is combined and multimodal data analysis technology is used to comprehensively evaluate the gasoline molecular structure corresponding to the characteristic peak information to generate a plurality of candidate gasoline molecular structures.

6. The method according to claim 1, characterized in that The step of simulating the interaction forces and energy states of the plurality of candidate gasoline molecular structures under different environmental parameters to obtain a preliminary gasoline molecular combination of the gasoline sample to be tested includes: Inputting the plurality of candidate gasoline molecular structures into quantum chemical calculation software, and setting a series of environmental parameters representing actual working conditions, including at least temperature and pressure; Using quantum chemical calculation methods, the interaction force and energy state of each candidate gasoline molecule structure under different environmental parameters are simulated, and the lowest energy conformation and stability of each gasoline molecule are calculated as the simulation result; Based on the simulation results, the gasoline molecular structures with the lowest energy and stability under the set environmental parameters are screened out to form a preliminary gasoline molecular structure list; Based on the preliminary gasoline molecular structure list, evaluate the relative stability of different gasoline molecular structures in the mixture and the distribution probability under different conditions; According to the relative stability and the distribution probability, the proportion of each gasoline molecule in the preliminary gasoline molecule combination is determined to generate the preliminary gasoline molecule combination of the gasoline sample to be tested.

7. A gasoline molecular composition prediction system, characterized in that: include: An acquisition module, used for acquiring infrared spectrum data and Raman spectrum data of the gasoline sample to be tested; An analysis module is used to analyze and process the characteristic peak information in the infrared spectrum data and the Raman spectrum data to obtain a plurality of candidate gasoline molecular structures of the gasoline sample to be tested, wherein the gasoline molecular structure refers to the specific arrangement of a single gasoline molecule; a simulation module, for simulating the interaction forces and energy states of the plurality of candidate gasoline molecular structures under different environmental parameters to obtain a preliminary gasoline molecular combination of the gasoline sample to be tested, wherein the gasoline molecular combination refers to a mixture composed of a plurality of different gasoline molecular structures; A prediction module, for evaluating the physical properties of each gasoline molecule in the preliminary gasoline molecular combination under specific conditions by using molecular dynamics simulation technology, and analyzing the relationship between the physical properties of each gasoline molecule and the corresponding gasoline molecular structure in combination with a pre-established prediction model, to predict the target gasoline molecular combination of the gasoline sample to be tested; An adjustment module, for adjusting the model parameters of the prediction model based on the comparison result between the target gasoline molecular combination and the gasoline molecular combination of the standard gasoline sample, so as to ensure the accuracy of the predicted target gasoline molecular combination, and generate a prediction report of the gasoline molecular combination of the gasoline sample to be tested; The molecular dynamics simulation technology is used to evaluate the physical properties of each gasoline molecule in the preliminary gasoline molecule combination under specific conditions, including: Inputting the preliminary gasoline molecular combination into molecular dynamics simulation software, and setting a series of environmental parameters representing actual working conditions, including at least temperature and pressure; Under the environmental parameters, simulating the dynamic behavior of each gasoline molecule in the preliminary gasoline molecule combination at a microscopic level to calculate the physical properties of each gasoline molecule, wherein the dynamic behavior at least includes the interaction, diffusion process, and aggregation morphology between gasoline molecules; Extracting key indicators from the physical properties of each gasoline molecule to generate an evaluation result of the physical properties of each gasoline molecule; The method of simulating the dynamic behavior of each gasoline molecule in the preliminary gasoline molecule combination at a microscopic level under the environmental parameters to calculate the physical properties of each gasoline molecule includes: Under the environmental parameters, the molecular dynamics simulation system is initialized to assign an initial position and an initial velocity to each gasoline molecule; Using molecular dynamics simulation software, simulating the dynamic behavior of each gasoline molecule in the preliminary gasoline molecule combination at the microscopic level, and in the simulation process, recording the position and speed of each gasoline molecule over time to obtain dynamic behavior data; Based on the dynamic behavior data, calculating the physical properties of each gasoline molecule, the physical properties at least including diffusion coefficient, viscosity, and density; Analyzing the relationship between the physical properties of each gasoline molecule and the corresponding gasoline molecular structure in combination with a pre-established prediction model, predicting the target gasoline molecule combination of the gasoline sample to be tested, including: Matching the physical property with the gasoline molecular structure of each gasoline molecule in the preliminary gasoline molecular combination to form a multidimensional feature data set including gasoline molecular structure and physical properties; Based on the multidimensional feature data set, using a pre-established prediction model, the relationship between the physical properties of each gasoline molecule and the corresponding gasoline molecular structure is analyzed to establish a mapping relationship model between the physical properties and the gasoline molecular structure; By using the mapping relationship model, each gasoline molecule in the preliminary gasoline molecule combination is evaluated to screen out gasoline molecules that exhibit the best physical properties under specific conditions; The screened gasoline molecules are combined to form a target gasoline molecule combination of the gasoline sample to be tested, ensuring that the target gasoline molecule combination has optimal physical properties under specific conditions.

Images