Prediction method and device for gasoline Reid vapor pressure, medium and equipment

Abstract

The invention discloses a gasoline Reid vapor pressure prediction method and device, a medium and equipment. The method comprises the steps that a gasoline sample to be predicted is obtained, gasoline fingerprints of the gasoline sample are determined, and the gasoline fingerprints comprise molecular components of the gasoline sample; inputting the gasoline fingerprint into a gasoline Reid vapor pressure prediction model to obtain the gasoline Reid vapor pressure of the to-be-predicted gasoline sample; the prediction model of the gasoline Reid vapor pressure is obtained by training a neural network model by using training data; the training data comprises gasoline fingerprints of the preset type of gasoline sample and gasoline Reid vapor pressure corresponding to the gasoline fingerprints, and the gasoline Reid vapor pressure is obtained by processing the gasoline fingerprints of the preset type of gasoline sample by using a gasoline Reid vapor pressure mechanism model. The technical problem that the model is repeatedly trained and does not have universality is solved, meanwhile, the prediction accuracy of the gasoline Reid vapor pressure is further improved, and the efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of crude oil processing technology, and in particular to a method, apparatus, medium, and equipment for predicting the Reid vapor pressure of gasoline. Background Technology

[0002] Reid vapor pressure (RVP) of gasoline is an important indicator of its evaporative performance, reflecting the tendency of fuel to vaporize in the engine during startup and when the engine is operating at high temperatures or latitudes. When the Reid vapor pressure is too low, it indicates that the fuel is difficult to vaporize, leading to difficulty starting the car, uneven mixing of fuel and air, incomplete combustion, and increased emissions of pollutants such as CO and hydrocarbons. Conversely, when the Reid vapor pressure is too high, it indicates excessively strong evaporative performance, causing the gasoline to vaporize easily and resulting in increased evaporative emissions, i.e., increased VOC (Volatile Organic Compounds) emissions. It can also create vapor lock in the fuel lines, interrupting fuel supply.

[0003] In the prediction of gasoline Red vapor pressure, the correlation method is used to correlate the infrared spectrum, gas chromatography, distillation curves and Red vapor pressure (RVP) of various gasoline samples to obtain a prediction model of gasoline Red vapor pressure. The prediction model of gasoline Red vapor pressure is then used to predict the Red vapor pressure of the oil product to be predicted. Summary of the Invention

[0004] A predictive model for gasoline Reid vapor pressure (RVP) obtained using a correlational method is used to predict the RVP of gasoline products. However, this model relies on an existing database storing infrared spectra, gas chromatography, distillation curves, and RVP data for various gasoline samples. For gasoline samples not stored in the database, a new database containing these data is required. Consequently, the resulting predictive model lacks universality, has high maintenance costs, and suffers from low efficiency and inaccuracy in predicting gasoline RVP.

[0005] In view of the above problems, the present invention is proposed to provide a method, apparatus, medium, or device for predicting gasoline Reid vapor pressure that overcomes or at least partially solves the above problems.

[0006] This embodiment provides a method for predicting the Reid vapor pressure of gasoline, including:

[0007] Obtain a gasoline sample to be predicted, and determine the gasoline fingerprint of the gasoline sample, wherein the gasoline fingerprint includes the molecular components of the gasoline sample;

[0008] The gasoline fingerprint is input into the gasoline Reid vapor pressure prediction model to obtain the gasoline Reid vapor pressure of the gasoline sample to be predicted. The gasoline Reid vapor pressure prediction model is obtained by training a neural network model using training data. The training data is obtained using an established gasoline Reid vapor pressure mechanism model. The training data includes gasoline fingerprints of preset type gasoline samples and gasoline Reid vapor pressures corresponding to the gasoline fingerprints. The gasoline Reid vapor pressure mechanism model is a calculation model obtained by simulating the experimental detection process of gasoline Reid vapor pressure and using the gas-liquid phase equilibrium principle in the experimental detection process.

[0009] A further optional implementation method is to obtain a predictive model for the Reid vapor pressure of gasoline in the following manner;

[0010] Training data were obtained using the established gasoline Reid vapor pressure mechanism model;

[0011] The preset training data is divided into a training set and a test set; the training set is input into the neural network model, and the hyperparameters of the neural network model are adjusted according to the loss function value of the neural network model. If the loss function value is not greater than the preset value, the trained neural network model is obtained.

[0012] The test set is input into the trained neural network model, and the output result is compared with the Reid vapor pressure of the test set data to obtain the prediction rate. If the prediction rate is less than the expected prediction rate, the trained neural network model is trained again until the prediction rate is not less than the expected prediction rate, and then the prediction model of gasoline Reid vapor pressure is obtained.

[0013] A further optional implementation involves obtaining training data using an established gasoline Reid vapor pressure mechanism model, including:

[0014] Constraints for virtual gasoline products are generated based on the molecular composition data of known types of gasoline samples, and gasoline samples of a preset type are randomly generated based on the constraints.

[0015] The gasoline fingerprint of a preset type of gasoline sample is processed using a gasoline Reid vapor pressure mechanism model to obtain the gasoline Reid vapor pressure of the preset type of gasoline sample. The gasoline fingerprint and the corresponding gasoline Reid vapor pressure of the preset type of gasoline sample constitute the training data.

[0016] A further optional implementation involves generating constraints based on molecular composition data of known types of gasoline samples, and randomly generating preset types of gasoline samples based on the constraints, including:

[0017] Based on the molecular types and molecular masses of known types of gasoline samples, the range of molecular types and molecular masses of each type of component is obtained.

[0018] Within the range of molecular types and molecular masses of molecular components, molecular types and molecular mass fractions of each type are randomly selected and normalized to obtain gasoline samples of the preset type.

[0019] In a further optional implementation, the gasoline fingerprint of a preset type of gasoline sample is processed using a gasoline Reid vapor pressure mechanism model to obtain the gasoline Reid vapor pressure of the preset type of gasoline sample. The specific process is as follows.

[0020] For each preset type of gasoline sample, the initial flash vapor pressure test experiment of gasoline based on the gasoline Red vapor pressure mechanism model is used to determine the initial flash model calculation conditions, which include operating temperature, first vaporization fraction, and operating pressure. Under the calculation conditions, the molecular components of the gasoline sample are processed using the flash model, and the second vaporization fraction and current operating pressure are obtained under the flash equilibrium state.

[0021] The difference in gasification fraction is obtained based on the first gasification fraction and the second gasification fraction.

[0022] If the difference in vaporization fraction is less than a preset accuracy threshold, the current operating pressure will be used as the Reid vapor pressure of the gasoline sample.

[0023] If the difference in vaporization fraction is not less than the preset accuracy threshold, the value of the second vaporization fraction is used as the value of the first vaporization fraction, and the steps of obtaining the second vaporization fraction and the current operating pressure are repeated until the difference in vaporization fraction is less than the preset accuracy threshold, so as to obtain the Reid vapor pressure of the gasoline sample, and then obtain the Reid vapor pressure of the gasoline sample of the preset type.

[0024] In a further optional implementation, under the stated calculation conditions, the molecular components of the gasoline sample are processed using a flash model. The flash model obtains the second vaporization fraction and the current operating pressure at flash equilibrium, including:

[0025] Under the aforementioned calculation conditions, the molecular components of the gasoline sample were processed using a flash evaporation model to obtain the gas phase content and liquid phase content of the gasoline sample.

[0026] The gas-liquid difference is obtained by comparing the gas phase content and the liquid phase content, and the state of the flash evaporation model is obtained by comparing the gas-liquid difference with the preset equilibrium threshold.

[0027] If the flash model is in flash equilibrium, the flash model calculates the second vaporization fraction of the molecular components of the gasoline sample.

[0028] If the flash model is in a non-flash equilibrium state, the operating pressure value is adjusted using a preset algorithm, and the process of determining whether a flash equilibrium state has been reached continues until the flash model is in a flash equilibrium state, at which point the second gasification fraction is calculated.

[0029] In a further optional implementation, under the stated calculation conditions, the molecular components of the gasoline sample are processed using a flash evaporation model to obtain the gas phase content and liquid phase content of the gasoline sample, including:

[0030] For each molecular component of the gasoline sample, under the aforementioned calculation conditions, a flash evaporation model is used to process one molecular component of the gasoline sample to obtain the gas phase fugacity coefficient and liquid phase fugacity coefficient of that molecular component.

[0031] Based on the gas phase fugacity coefficient and the liquid phase fugacity coefficient, the phase equilibrium constant of one molecular component of the gasoline sample is obtained;

[0032] Based on the phase equilibrium constant, the first vaporization fraction, and a molecular component, the gas phase content and liquid phase content of a molecular component of the gasoline sample are calculated, thereby obtaining the gas phase content and liquid phase content of the gasoline sample.

[0033] In a further optional implementation, the gas-liquid difference is obtained by comparing the gas phase content and the liquid phase content, and the state of the flash evaporation model is obtained by comparing the gas-liquid difference with a preset equilibrium threshold, including:

[0034] The difference between the gas phase content and the liquid phase content is compared to obtain the gas-liquid difference value;

[0035] The gas-liquid difference is compared with a preset equilibrium threshold. If the gas-liquid difference is less than the preset equilibrium threshold, the flash evaporation model is in a flash equilibrium state. If the gas-liquid difference is not less than the preset equilibrium threshold, the flash evaporation model is in a non-flash equilibrium state.

[0036] In a further optional implementation, if the flash model is in flash equilibrium, the flash model calculates a second vaporization fraction from the molecular components of the gasoline sample, including:

[0037] If the flash model is in flash equilibrium, the density of the mixed gas phase is obtained by calculating the molecular composition of the gasoline sample using the Soaf-Redlich-Kuang equation, and the density of the mixed liquid phase is obtained by calculating the molecular composition of the gasoline sample using the Rector equation; based on the density of the mixed gas phase and the density of the mixed liquid phase, the second vaporization fraction is obtained.

[0038] A further optional implementation method involves adjusting the operating pressure value using a preset algorithm, including:

[0039] The operating pressure value is adjusted using Newton's iteration method.

[0040] A further optional implementation involves adjusting the hyperparameters of the neural network model based on the loss function value of the neural network model, including:

[0041] Based on the loss function value of the neural network model, adjust at least one of the following: the number of hidden layers, the number of nodes in the hidden layers, the learning rate, and the number of iterations.

[0042] This invention provides a device for predicting the red vapor pressure of gasoline, comprising:

[0043] The data acquisition module is used to acquire the gasoline sample to be predicted and determine the gasoline fingerprint of the gasoline sample, wherein the gasoline fingerprint includes the molecular components of the gasoline sample;

[0044] The prediction module is used to input the gasoline fingerprint into the gasoline Reid vapor pressure prediction model to obtain the gasoline Reid vapor pressure of the gasoline sample to be predicted. The gasoline Reid vapor pressure prediction model is obtained by training a neural network model using training data. The training data is obtained using an established gasoline Reid vapor pressure mechanism model. The training data includes gasoline fingerprints of preset type gasoline samples and gasoline Reid vapor pressures corresponding to the gasoline fingerprints. The gasoline Reid vapor pressure mechanism model is a calculation model obtained by simulating the experimental detection process of gasoline Reid vapor pressure and using the gas-liquid phase equilibrium principle in the experimental detection process.

[0045] This invention provides a computer storage medium storing computer-executable instructions, which, when executed by a processor, implement the gasoline red vapor pressure prediction method described above.

[0046] This invention provides a terminal device for predicting gasoline Red vapor pressure, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the above-described method for predicting gasoline Red vapor pressure.

[0047] The beneficial effects of the above-described technical solutions provided in the embodiments of the present invention include at least the following:

[0048] The prediction model for gasoline Reid vapor pressure is obtained by training a neural network model using training data, which includes gasoline fingerprints of a preset type of gasoline sample and the corresponding gasoline Reid vapor pressure. The gasoline Reid vapor pressure is obtained by processing the gasoline fingerprints of the preset type of gasoline sample using a gasoline Reid vapor pressure mechanism model. The preset type of gasoline sample includes a complete range of gasoline samples, and the prediction model trained using this data does not require repeated training, overcoming the technical problems of repeated model training and lack of universality. At the same time, it further improves the prediction accuracy of gasoline Reid vapor pressure and increases efficiency.

[0049] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0050] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0051] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0052] Figure 1 This is a flowchart of the method for predicting the red vapor pressure of gasoline in an embodiment of the present invention;

[0053] Figure 2 This is a flowchart illustrating the calculation of the gasoline red vapor pressure of a preset type of gasoline sample in an embodiment of the present invention;

[0054] Figure 3 This is a flowchart illustrating the calculation of the second gasification fraction and the current operating pressure in an embodiment of the present invention;

[0055] Figure 4 This is a schematic diagram of the structure of the gasoline Red vapor pressure prediction device in an embodiment of the present invention;

[0056] Figure 5 This is a complete flowchart of the method for predicting the red vapor pressure of gasoline in an embodiment of the present invention. Detailed Implementation

[0057] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0058] To address the problems existing in the prior art, embodiments of the present invention provide a method, apparatus, medium, and device for predicting the red vapor pressure of gasoline.

[0059] This invention provides a method for predicting the Reid vapor pressure of gasoline, the process of which is as follows: Figure 1 As shown, it includes the following steps:

[0060] Step S101: Obtain the gasoline sample to be predicted and determine the gasoline fingerprint of the gasoline sample. The gasoline fingerprint includes the molecular components of the gasoline sample.

[0061] In this embodiment, a gasoline sample to be predicted is obtained, and the gasoline fingerprint of the gasoline sample to be predicted is determined by gas chromatography. Depending on the application scenario, other methods can also be used to determine the gasoline fingerprint of the gasoline sample to be predicted; wherein the gasoline fingerprint includes the molecular components of the gasoline sample.

[0062] It should be noted that gasoline fingerprinting refers to the unique characteristics reflected by specific chemical components and proportions in gasoline resources. Gasoline fingerprinting can be used to identify and distinguish different types of gasoline. Chemical descriptors can be used to identify the molecules contained in a gasoline sample, and the molecular components of the gasoline sample are used as its gasoline fingerprint.

[0063] Gasoline is a very complex mixture. The same type of gasoline may contain gasoline molecules of different sizes and dimensions. Gasoline fingerprints can be used to map gasoline samples of the same type with different numbers of molecules to a gasoline database containing a certain type of gasoline with a fixed number of molecules.

[0064] Step S102: Input the gasoline fingerprint into the gasoline Reid vapor pressure prediction model to obtain the gasoline Reid vapor pressure of the gasoline sample to be predicted.

[0065] In this embodiment, the prediction model for gasoline Reid vapor pressure is obtained by training a neural network model using training data; the training data is obtained using an established gasoline Reid vapor pressure mechanism model; the gasoline Reid vapor pressure mechanism model is a calculation model obtained by simulating the experimental detection process of gasoline Reid vapor pressure and using the gas-liquid phase equilibrium principle in the experimental detection process.

[0066] Specifically, training data is obtained using the established gasoline Reid vapor pressure mechanism model. This training data is then used to train a neural network model to obtain a predictive model for gasoline Reid vapor pressure. For example, constraints for virtual gasoline products are generated based on the molecular composition data of known types of gasoline samples. Preset types of gasoline samples are then randomly generated based on these constraints. The gasoline fingerprints of the preset types of gasoline samples are processed using the gasoline Reid vapor pressure mechanism model to obtain their corresponding gasoline Reid vapor pressures. These gasoline fingerprints and their corresponding gasoline Reid vapor pressures constitute the training data.

[0067] The process of obtaining a preset type of gasoline sample is as follows: Constraints for a virtual gasoline product are generated based on the molecular component data of a known type of gasoline sample; then, a preset type of gasoline sample is randomly generated based on these constraints. For example, based on the molecular types and molecular masses of the known type of gasoline sample, the range of molecular types and molecular masses for each molecular component is obtained; within this range, the molecular types and molecular mass fractions are randomly selected and normalized to obtain the preset type of gasoline sample.

[0068] Specifically, gasoline samples are complex mixtures, mainly composed of various hydrocarbon groups, including straight-chain alkanes, branched-chain alkanes, cycloalkanes, and aromatics. The proportions and types of these different hydrocarbons in gasoline vary depending on their source, refining process, and intended use. Based on the known PIONA composition of gasoline types, the generation conditions for the molecular types of a preset type of gasoline sample are given. Based on the range of molecular mass fractions of various groups in known types of gasoline, the generation conditions for the molecular mass fractions of various groups in the preset type of gasoline sample are determined. Under these conditions, the molecular types and molecular mass fractions of the preset type of gasoline sample are randomly generated, and finally normalized to obtain a wide variety of preset type gasoline samples. It should be noted that the molecular components of a preset type of gasoline sample contain multiple molecular components of different types. Each molecular component of a type contains one molecule of that type and its corresponding molecular mass fraction. For example, the molecular components of type A gasoline include type A molecular components and type B molecular components. Type A molecular components contain type A molecules and their corresponding molecular mass fractions, and type B molecular components contain type B molecules and their corresponding molecular mass fractions.

[0069] The process of obtaining training data is as follows: Gasoline Reid vapor pressure of gasoline samples of a preset type is obtained by processing the gasoline fingerprint using the gasoline Reid vapor pressure mechanism model. See details below. Figure 2 The training data consisted of gasoline fingerprints and corresponding gasoline red vapor pressures of preset type gasoline samples.

[0070] The training process of the neural network model is as follows: The training data is divided into a training set and a test set; the training set is input into the neural network model, and the hyperparameters of the neural network model are adjusted according to the loss function value. If the loss function value is not greater than a preset value, a trained neural network model is obtained; the test set is input into the trained neural network model, and the output result is compared with the Reid vapor pressure of the test set data to obtain the prediction rate; if the prediction rate is less than the expected prediction rate, the trained neural network model is trained again until the prediction rate is not less than the expected prediction rate, thus obtaining the prediction model for gasoline Reid vapor pressure. The training set is used to optimize the network structure parameters of the neural network model, and the test set is used to evaluate the generalization performance of the trained neural network model.

[0071] For example, the training data can be divided into a training set and a test set. The training set is used to optimize the network structure parameters of the initial Reid vapor pressure prediction model, while the test set is used to evaluate the generalization performance of the trained Reid vapor pressure prediction model. Gasoline fingerprints from the test set are input into the initial Reid vapor pressure prediction model to obtain the corresponding predicted Reid vapor pressure values. A loss function is constructed, for example, using the root mean square error function. The loss value is calculated based on the predicted value and the gasoline Reid vapor pressure from the test set. An optimization algorithm, such as gradient descent, is used to continuously adjust the hyperparameters of the neural network, such as adjusting the number of hidden layers and nodes, to reduce the loss value until the model converges, thus obtaining the Reid vapor pressure prediction model.

[0072] In this embodiment, experimental methods are used to obtain more realistic Reid vapor pressures of gasoline samples of a preset type. Based on the molecular composition of the preset type of gasoline samples and the gasoline Reid vapor pressures corresponding to gasoline fingerprints, training data is obtained to train a Reid vapor pressure prediction model with higher accuracy. For example, gas chromatography is used to detect known types of gasoline samples, including common middle fractions and finished gasoline in gasoline blending tanks, to obtain qualitative and quantitative information on the real gasoline molecules. Upper and lower limits are added to the mass fraction of each molecule of real gasoline, for example, the upper limit is set to 130% of the mass fraction of each molecule, and the lower limit is set to 70% of the mass fraction of each molecule. Molecular composition data of preset type gasoline samples are randomly generated within this range. The Reid vapor pressure prediction model does not require repeated training, overcoming the technical problems of repeated model training and lack of universality, while further improving the prediction accuracy of gasoline Reid vapor pressure and increasing efficiency.

[0073] This embodiment, based on known types of gasoline samples, derives a rule for generating preset types of gasoline samples, and generates preset types of gasoline samples according to this rule. The preset types of gasoline samples are derived from known types of gasoline samples, and can reflect real gasoline products, thus expanding the total variety of gasoline products. Based on the preset types of gasoline samples, gasoline fingerprints are obtained. Using a gasoline Reid vapor pressure mechanism model, the gasoline fingerprints are processed to obtain the Reid vapor pressure of the gasoline samples. The gasoline fingerprints and their corresponding Reid vapor pressures provide sufficient data for subsequent model training and validation. The gasoline fingerprints and their corresponding Reid vapor pressures are used to train and validate the neural network model. The gasoline Reid vapor pressure prediction model establishes a close correlation between the gasoline fingerprints and their corresponding Reid vapor pressures. The gasoline Reid vapor pressure prediction model processes the gasoline fingerprints of the gasoline samples to be predicted, accurately obtaining the Reid vapor pressure density of the gasoline samples to be predicted.

[0074] In this embodiment, the gasoline Reid vapor pressure of a preset type of gasoline sample is obtained by processing the gasoline fingerprint using a gasoline Reid vapor pressure mechanism model. The specific process is as follows: Figure 2 As shown, the process includes the following steps:

[0075] Step S201: For each preset type of gasoline sample, determine the initial flash model calculation conditions based on the Reid vapor pressure test experiment of gasoline according to the gasoline Reid vapor pressure mechanism model. The initial flash model calculation conditions include operating temperature, first vaporization fraction, and operating pressure. Under the initial flash model calculation conditions, process the molecular components of the gasoline sample using the flash model. The flash model obtains the second vaporization fraction and the current operating pressure under the flash equilibrium state.

[0076] In this embodiment, as Figure 5 The flash model of the gasoline Reid vapor pressure mechanism model, as shown, calculates the molecular composition of the gasoline sample under determined initial flash model calculation conditions, including operating temperature, first vaporization fraction, and operating pressure, to obtain the second vaporization fraction when the flash model reaches equilibrium. For example, based on the Reid vapor pressure test experiment of gasoline, the initial flash model calculation conditions are determined to be an operating temperature of 37.8℃, and the first vaporization fraction and operating pressure. It should be noted that the operating temperature, first vaporization fraction, and operating pressure may vary depending on the environment. Under the initial flash model calculation conditions of operating temperature, first vaporization fraction, and operating pressure, the flash model calculates the molecular composition of the gasoline sample, and the flash model obtains the second vaporization fraction and the current operating pressure at flash equilibrium.

[0077] The process of obtaining the second gasification fraction and the current operating pressure under the flash evaporation equilibrium state in the flash evaporation model is described in [reference needed]. Figure 3 The process in the middle.

[0078] Step S202: Based on the first vaporization fraction and the second vaporization fraction, obtain the difference in vaporization fraction. Calculate the Reid vapor pressure of gasoline based on the difference in vaporization fraction, and then obtain the Reid vapor pressure of the preset type of gasoline sample.

[0079] In this embodiment, as Figure 5 The calculation yields the second vaporization fraction and the current operating pressure. The difference between the vaporization fractions and a preset accuracy threshold is compared. If they match, the specific Reid vapor pressure of the gasoline corresponding to the molecular component is obtained; otherwise, the first vaporization fraction and operating pressure are adjusted. A specific flash vaporization model mathematically calculates the difference between the first and second vaporization fractions. If the difference is less than the preset accuracy threshold, the current operating pressure is used as the Reid vapor pressure of the gasoline sample. If the difference is not less than the preset accuracy threshold, the value of the second vaporization fraction is used as the value of the first vaporization fraction, and the steps to obtain the second vaporization fraction and the current operating pressure are repeated until the difference is less than the preset accuracy threshold, thus obtaining the Reid vapor pressure of the gasoline sample, and subsequently, the Reid vapor pressure of a gasoline sample of a preset type.

[0080] The preset accuracy threshold is set according to different environments. For example, if the preset accuracy threshold is set to 0.0001, when the vaporization fraction difference is less than 0.0001, the current operating pressure is used as the Reid vapor pressure of the gasoline sample; when the vaporization fraction difference is not less than 0.0001, the value of the second vaporization fraction is used as the value of the first vaporization fraction, and the steps of obtaining the second vaporization fraction and the current operating pressure are repeated until the vaporization fraction difference is less than the preset accuracy threshold, thus obtaining the Reid vapor pressure of the gasoline sample, and then obtaining the Reid vapor pressure of the preset type of gasoline sample. In the step of re-executing the steps of obtaining the second vaporization fraction and the current operating pressure, the current operating pressure is adjusted using the Newton-Raphson iteration method, specifically adjusting the value of the current operating pressure to the value of the previous operating pressure.

[0081] A further optional implementation involves obtaining the second vaporization fraction and the current operating pressure using the flash model at flash equilibrium, as follows: Figure 3 As shown, the process includes the following steps:

[0082] Step S301: Under the initial flash model calculation conditions, the molecular components of the gasoline sample are processed using the flash model to obtain the gas phase content and liquid phase content of the gasoline sample.

[0083] In this embodiment, SMILES descriptor, molecular weight, critical temperature, critical pressure, eccentricity factor, liquid phase molar volume at 25°C, solubility parameters, etc., can also be obtained. The obtained parameters are used to process the molecular components of the gasoline sample using a flash evaporation model to obtain the gas phase content and liquid phase content of the gasoline sample.

[0084] Among them, the fugacity coefficients of the molecular components of the gasoline sample are calculated as follows: Figure 5 As shown, for each molecular component of the gasoline sample, under the initial flash evaporation model calculation conditions, the flash evaporation model is used to process one molecular component of the gasoline sample to obtain the gas-phase fugacity coefficient and liquid-phase fugacity coefficient of that molecular component. For example, the gas-phase fugacity coefficient φ of one molecular component of the gasoline sample is calculated by simultaneously solving the SRK (Soave-Redlich-Kwong) equations of state. i v and liquid phase fugacity coefficient φ i L The specific calculation formula is as follows: and Z i Let be the compressibility factor of the i-th molecule in the preset type of oil sample, T be the operating temperature, R be the gas constant, and q = (Ψ*a(T)). r )) / (Ω*T r The parameters σ, ∈, a(T), ψ, Ω, etc. in the formula are obtained from the SRK state equations and parameter tables in Table 1; β=Ω*(P r / T r ), P r =P / P c T r =T / T c T c P is the critical temperature. c Where P is the critical pressure, a is the operating pressure, and P is the critical pressure. SRK (T r ) represents the temperature function of the Soave-Redlich-Kwong equation, a PR (T r ) is the temperature function of the Peng-Robinson equation, and ω is the eccentricity factor.

[0085]

[0086]

[0087] Table 1. SRK State Equations and Parameters

[0088] Calculate the gas phase and liquid phase contents of the molecular components of a gasoline sample as follows: Figure 5 As shown, the phase equilibrium constant of a molecular component of a gasoline sample is obtained based on the gas phase fugacity coefficient and the liquid phase fugacity coefficient. Based on the phase equilibrium constant, the first vaporization fraction, and the molecular component, the gas phase content and liquid phase content of the molecular component of the gasoline sample are calculated, and thus the gas phase content and liquid phase content of the gasoline sample are obtained.

[0089] Calculate the phase equilibrium constants of the molecular components of the gasoline sample as follows: Figure 5 As shown, based on the calculated gas-phase fugacity coefficient and liquid-phase fugacity coefficient of each molecular component of the preset type gasoline sample, the phase equilibrium constant of each molecular component of the preset type gasoline sample is determined. For example, using... The formula calculates the phase equilibrium constant, K, of any molecular component in a gasoline sample of a preset type. i The phase equilibrium constant of the i-th molecular component of a gasoline sample of a preset type. The gas phase fugacity coefficient of the i-th molecular component in a predefined type of gasoline sample. is the liquid phase fugacity coefficient of the i-th molecular component in a pre-defined gasoline sample.

[0090] use and The formula calculates the phase equilibrium constant of each molecular component of the predicted type gasoline sample, the first vaporization fraction of the flash model, and the molecular components, as well as the gas phase content xi and liquid phase content yi of any component of the preset type gasoline sample. Wherein, Z i K represents the mole fraction of the i-th molecular component in a predefined type of gasoline sample. i Let be the phase equilibrium constant of the i-th molecular component of the preset type of gasoline sample, and e be the first vaporization fraction.

[0091] Step S302: The gas-liquid difference is obtained by comparing the gas phase content and the liquid phase content. The gas-liquid difference is compared with the preset equilibrium threshold to obtain the state of the flash model. If the state of the flash model is the flash equilibrium state, the flash model calculates the second vaporization fraction of the molecular components of the gasoline sample. If the state of the flash model is not the flash equilibrium state, the operating pressure value is adjusted using the preset algorithm, and the step of whether the flash equilibrium state has been reached is continued until the state of the flash model is the flash equilibrium state, and the second vaporization fraction is calculated.

[0092] In this embodiment, as Figure 5 The gas-liquid difference is compared with a preset equilibrium threshold. If it does not meet the threshold, the operating pressure is adjusted; if it does meet the threshold, the second vaporization fraction and the current operating pressure are calculated. Specifically, the difference between the gas phase content and the liquid phase content is compared to obtain the gas-liquid difference value. This gas-liquid difference value is then compared with a preset equilibrium threshold. If the gas-liquid difference value is less than the preset equilibrium threshold, the flash evaporation model is in flash equilibrium; if the gas-liquid difference value is not less than the preset equilibrium threshold, the flash evaporation model is in non-flash equilibrium. The preset algorithm is Newton's iteration method, but other iterative methods can be used depending on the environment.

[0093] For example, based on the phase equilibrium constants of each molecular component of the preset type of gasoline sample obtained from the above calculations, as well as the gas phase content xi and liquid phase content yi, the Rachford-Rice equation is established, and the calculation formula of this equation is used. The gas-liquid difference is calculated, and if it is less than a preset equilibrium threshold, the preset equilibrium threshold is set according to different environments. For example, if the equilibrium threshold is set to 0.001, the flash evaporation model is determined to have reached flash equilibrium when the gas-liquid difference is less than 0.0001; otherwise, the flash evaporation model is determined not to have reached flash equilibrium when the gas-liquid difference is not less than the preset equilibrium threshold, i.e., not less than 0.0001, and the operating pressure value is adjusted using Newton's iteration method.

[0094] In this embodiment, under flash equilibrium conditions, the molecular composition of the gasoline sample is calculated using the Soaf-Redlich-Kuang equation to obtain the density of the mixed gas phase, and the molecular composition of the gasoline sample is calculated using the Rackett equation to obtain the density of the mixed liquid phase. Based on the densities of the mixed gas and mixed liquid phases, the second vaporization fraction is obtained. For example, the density of the mixed gas phase separated when the flash model reaches flash equilibrium can be calculated using the SRK equation of state for a preset type of gasoline sample; the density of the mixed liquid phase separated when the flash model reaches flash equilibrium can be calculated using the improved Rackett equation; and the second vaporization fraction is calculated based on the densities of the mixed gas and mixed liquid phases. Specifically, the utilization of… The molar volume V of the mixed gas phase is obtained by formula. mix V Using the formula Calculate the density of the mixed gas phase. Z is the compressibility factor, which can be obtained based on the molecular composition of a preset type of gasoline sample; R is the gas constant; T is the operating temperature; P is the operating pressure; MW i Let X be the molecular mass of the i-th molecule in the mixed gas phase. i Let Z be the mass content of the i-th molecule in the mixed gas phase; where the compressibility factor Z is calculated using B0 = 0.083 - 0.422 / (Tr^ 1.6 B1 = 0.139 - 0.172 / (Tr^ 4 ·2), B=R*Tc·*(B0+ω· ★ B1) / Pc, Z=1+B· ★ The formula P· / (R*T) is used to calculate the value.

[0095] Among them, utilizing The formula calculates the molar volume of each molecular component in the mixed liquid phase, and then, based on the molecular weight of each molecular component, the density ρ of each molecular component in the mixed liquid phase is calculated. i;use Formula for calculating the density ρ of the mixed liquid phase mix Where Vs is the molar volume of any molecular component in the mixed liquid phase, Tc is the critical temperature of that molecular component, Pc is the critical pressure, T is the operating temperature, ω is the eccentricity factor, and x wi ρ represents the mass content of the i-th molecular component in the mixed liquid phase. i Let be the density of the i-th molecular component in the mixed liquid phase.

[0096] Among them, utilizing The formula calculates the second vaporization fraction of the gasoline sample of the preset type, where ρ LP The density of a pre-defined type of gasoline sample at 37.8℃; MW V The molecular weight of the mixed gas phase can be calculated based on the molecular composition of a gasoline sample of a preset type; MW L The molecular weight of the mixed liquid phase can be calculated based on the molecular composition of a gasoline sample of a preset type.

[0097] In the embodiments provided in this application, a gasoline Reid vapor pressure mechanism model is used to process the molecular components of a large number of preset types of gasoline samples. Training data is obtained based on the molecular components of the preset types of gasoline samples and their corresponding gasoline Reid vapor pressures. A Reid vapor pressure prediction model is then trained using this training data. Finally, the Reid vapor pressure prediction model is used to obtain the Reid vapor pressure of the gasoline sample based on its molecular component data. Since the gasoline Reid vapor pressure mechanism model simulates the experimental detection process of gasoline Reid vapor pressure and utilizes the gas-liquid phase equilibrium principle during the experimental detection process, its calculation results have high accuracy and can be applied to different types of gasoline. Therefore, using the gasoline Reid vapor pressure mechanism model, training data covering the Reid vapor pressure characteristics of various existing types of gasoline can be obtained accurately, conveniently, and at low cost. This training data can then be used to train a Reid vapor pressure prediction model that can quickly and accurately predict the Reid vapor pressure characteristics of gasoline samples. The Reid vapor pressure prediction model overcomes the problems of complex construction and long running time of the Reid vapor pressure mechanism model and can be widely used in industrial real-time optimization.

[0098] Based on the same inventive concept, embodiments of the present invention also provide a device for predicting the Reid vapor pressure of gasoline, the structure of which is as follows: Figure 4 As shown, it includes: a data acquisition module 401 and a prediction module 402;

[0099] The data acquisition module 401 is used to acquire the gasoline sample to be predicted and determine the gasoline fingerprint of the gasoline sample. The gasoline fingerprint includes the molecular components of the gasoline sample.

[0100] The prediction module 402 is used to input the gasoline fingerprint into the gasoline Reid vapor pressure prediction model to obtain the gasoline Reid vapor pressure of the gasoline sample to be predicted. The gasoline Reid vapor pressure prediction model is obtained by training a neural network model using training data. The training data is obtained using an established gasoline Reid vapor pressure mechanism model. The training data includes the gasoline fingerprint of a preset type of gasoline sample and the gasoline Reid vapor pressure corresponding to the gasoline fingerprint. The gasoline Reid vapor pressure mechanism model is a calculation model obtained by simulating the experimental detection process of gasoline Reid vapor pressure and using the gas-liquid phase equilibrium principle in the experimental detection process.

[0101] This invention also provides a computer storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method for predicting gasoline red vapor pressure.

[0102] This invention also provides a terminal device for predicting gasoline Red vapor pressure, characterized in that it includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the above-mentioned method for predicting gasoline Red vapor pressure.

[0103] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0104] Unless otherwise specifically stated, terms such as processing, calculation, operation, determination, display, etc., may refer to the actions and / or processes of one or more processing or computing systems or similar devices that represent the manipulation and conversion of data representing physical (e.g., electronic) quantities within the registers or memory of the processing system into other data similarly representing physical quantities within the memory, registers, or other such information storage, transmission, or display devices of the processing system. Information and signals can be represented using any of a variety of different techniques and methods. For example, data, instructions, commands, information, signals, bits, symbols, and chips mentioned throughout the above description can be represented by voltage, current, electromagnetic waves, magnetic fields or particles, light fields or particles, or any combination thereof.

[0105] It should be understood that the specific order or hierarchy of steps in the disclosed process is an example of an exemplary method. Based on design preferences, it should be understood that the specific order or hierarchy of steps in the process may be rearranged without departing from the scope of this disclosure. The appended method claims provide elements of various steps in an exemplary order and are not intended to limit the scope to the specific order or hierarchy described.

[0106] In the detailed description above, various features are combined together in a single embodiment to simplify this disclosure. This approach to disclosure should not be construed as reflecting an intention that embodiments of the claimed subject matter require more features than are explicitly stated in each claim. Rather, as reflected in the appended claims, the invention is presented with fewer features than all of the features in a single disclosed embodiment. Therefore, the appended claims are hereby explicitly incorporated into the detailed description, with each claim representing a separate preferred embodiment of the invention.

[0107] Those skilled in the art will also understand that the various illustrative logic blocks, modules, circuits, and algorithm steps described in conjunction with the embodiments herein can be implemented as electronic hardware, computer software, or a combination thereof. To clearly illustrate the interchangeability between hardware and software, the various illustrative components, blocks, modules, circuits, and steps described above are generally described in terms of their functionality. Whether such functionality is implemented as hardware or software depends on the specific application and the design constraints imposed on the overall system. Those skilled in the art can implement the described functionality in alternative ways for each specific application; however, such implementation decisions should not be construed as departing from the scope of this disclosure.

[0108] The steps of the methods or algorithms described in conjunction with the embodiments herein can be directly embodied in hardware, software modules executed by a processor, or a combination thereof. The software modules can reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium well known in the art. An exemplary storage medium is connected to the processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. The ASIC can reside in a user terminal. Alternatively, the processor and storage medium can exist as discrete components in the user terminal.

[0109] For software implementation, the techniques described in this application can be implemented using modules (e.g., procedures, functions, etc.) that perform the functions described in this application. This software code can be stored in memory units and executed by a processor. The memory units can be implemented within the processor or outside the processor; in the latter case, they are communicatively coupled to the processor via various means, as is well known in the art.

[0110] The foregoing description includes examples of one or more embodiments. It is certainly impossible to describe all possible combinations of components or methods in order to describe the above embodiments, but those skilled in the art will recognize that further combinations and arrangements of the various embodiments are possible. Therefore, the embodiments described herein are intended to cover all such changes, modifications, and variations that fall within the scope of the appended claims. Furthermore, the term "comprising" as used in the specification or claims is interpreted in a manner similar to the term "including," as interpreted when used as a conjunction in the claims. Additionally, the use of any term "or" in the specification of the claims is intended to mean "non-exclusive or."

Claims

1. A method for predicting the Reid vapor pressure of gasoline, characterized in that, include: Obtain a gasoline sample to be predicted, and determine the gasoline fingerprint of the gasoline sample, wherein the gasoline fingerprint includes the molecular components of the gasoline sample; The gasoline fingerprint is input into the gasoline Reid vapor pressure prediction model to obtain the gasoline Reid vapor pressure of the gasoline sample to be predicted. The gasoline Reid vapor pressure prediction model is obtained by training a neural network model using training data. The training data is obtained using an established gasoline Reid vapor pressure mechanism model. The training data includes gasoline fingerprints of preset type gasoline samples and gasoline Reid vapor pressures corresponding to the gasoline fingerprints. The gasoline Reid vapor pressure mechanism model is a calculation model obtained by simulating the experimental detection process of gasoline Reid vapor pressure and using the gas-liquid phase equilibrium principle in the experimental detection process.

2. The method as described in claim 1, wherein a prediction model for gasoline Reid vapor pressure is obtained in the following manner; Training data were obtained using the established gasoline Reid vapor pressure mechanism model; The training data is divided into a training set and a test set; the training set is input into the neural network model, and the hyperparameters of the neural network model are adjusted according to the loss function value of the neural network model. If the loss function value is not greater than a preset value, then a trained neural network model is obtained. The test set is input into the trained neural network model, and the output result is compared with the Reid vapor pressure of the test set data to obtain the prediction rate. If the prediction rate is less than the expected prediction rate, the trained neural network model is trained again until the prediction rate is not less than the expected prediction rate, and then the prediction model of gasoline Reid vapor pressure is obtained.

3. The method as described in claim 1, characterized in that, Training data was obtained using the established gasoline Reid vapor pressure mechanism model, including: Constraints for virtual gasoline products are generated based on the molecular composition data of known types of gasoline samples, and gasoline samples of a preset type are randomly generated based on the constraints. The gasoline fingerprint of a preset type of gasoline sample is processed using a gasoline Reid vapor pressure mechanism model to obtain the gasoline Reid vapor pressure of the preset type of gasoline sample. The gasoline fingerprint and the corresponding gasoline Reid vapor pressure of the preset type of gasoline sample constitute the training data.

4. The method as described in claim 3, characterized in that, Constraints are generated based on the molecular composition data of known types of gasoline samples. Preset types of gasoline samples are then randomly generated based on these constraints, including: Based on the molecular types and molecular masses of known types of gasoline samples, the range of molecular types and molecular masses of each type of component is obtained. Within the range of molecular types and molecular masses of molecular components, molecular types and molecular mass fractions of each type are randomly selected and normalized to obtain gasoline samples of the preset type.

5. The method as described in claim 4, characterized in that, The gasoline Reid vapor pressure of a preset type of gasoline sample is obtained by processing the gasoline fingerprint using the gasoline Reid vapor pressure mechanism model. The specific process is as follows: For each preset type of gasoline sample, the initial flash vapor pressure test experiment of gasoline based on the gasoline Red vapor pressure mechanism model is used to determine the initial flash model calculation conditions, which include operating temperature, first vaporization fraction, and operating pressure. Under the calculation conditions, the molecular components of the gasoline sample are processed using the flash model, and the second vaporization fraction and current operating pressure are obtained in the flash equilibrium state. The difference in gasification fraction is obtained based on the first gasification fraction and the second gasification fraction. If the difference in vaporization fraction is less than a preset accuracy threshold, the current operating pressure will be used as the Reid vapor pressure of the gasoline sample. If the difference in vaporization fraction is not less than the preset accuracy threshold, the value of the second vaporization fraction is used as the value of the first vaporization fraction, and the steps of obtaining the second vaporization fraction and the current operating pressure are repeated until the difference in vaporization fraction is less than the preset accuracy threshold, so as to obtain the Reid vapor pressure of the gasoline sample, and then obtain the Reid vapor pressure of the gasoline sample of the preset type.

6. The method as described in claim 5, characterized in that, Under the stated calculation conditions, the molecular components of the gasoline sample are processed using a flash evaporation model. The flash evaporation model, at flash evaporation equilibrium, yields the second vaporization fraction and the current operating pressure, including: Under the aforementioned calculation conditions, the molecular components of the gasoline sample were processed using a flash evaporation model to obtain the gas phase content and liquid phase content of the gasoline sample. The gas-liquid difference is obtained by comparing the gas phase content and the liquid phase content, and the state of the flash evaporation model is obtained by comparing the gas-liquid difference with the preset equilibrium threshold. If the flash model is in flash equilibrium, the flash model calculates the second vaporization fraction of the molecular components of the gasoline sample. If the flash model is in a non-flash equilibrium state, the operating pressure value is adjusted using a preset algorithm, and the process of determining whether a flash equilibrium state has been reached continues until the flash model is in a flash equilibrium state, at which point the second gasification fraction is calculated.

7. The method as described in claim 6, characterized in that, Under the aforementioned calculation conditions, the molecular components of the gasoline sample are processed using a flash evaporation model to obtain the gas phase content and liquid phase content of the gasoline sample, including: For each molecular component of the gasoline sample, under the aforementioned calculation conditions, a flash evaporation model is used to process one molecular component of the gasoline sample to obtain the gas phase fugacity coefficient and liquid phase fugacity coefficient of that molecular component. Based on the gas phase fugacity coefficient and the liquid phase fugacity coefficient, the phase equilibrium constant of one molecular component of the gasoline sample is obtained; Based on the phase equilibrium constant, the first vaporization fraction, and a molecular component, the gas phase content and liquid phase content of a molecular component of the gasoline sample are calculated, thereby obtaining the gas phase content and liquid phase content of the gasoline sample.

8. The method as described in claim 7, characterized in that, The gas-liquid difference is obtained by comparing the gas phase content and the liquid phase content, and the state of the flash evaporation model is obtained by comparing the gas-liquid difference with a preset equilibrium threshold, including: The difference between the gas phase content and the liquid phase content is compared to obtain the gas-liquid difference value; The gas-liquid difference is compared with a preset equilibrium threshold. If the gas-liquid difference is less than the preset equilibrium threshold, the flash evaporation model is in a flash equilibrium state. If the gas-liquid difference is not less than the preset equilibrium threshold, the flash evaporation model is in a non-flash equilibrium state.

9. The method as described in claim 8, characterized in that, If the flash model is in flash equilibrium, the flash model calculates the second vaporization fraction of the gasoline sample based on its molecular composition, including: If the flash model is in flash equilibrium, the density of the mixed gas phase is obtained by calculating the molecular composition of the gasoline sample using the Soaf-Redlich-Kuang equation, and the density of the mixed liquid phase is obtained by calculating the molecular composition of the gasoline sample using the Rector equation; based on the density of the mixed gas phase and the density of the mixed liquid phase, the second vaporization fraction is obtained.

10. The method as described in claim 6, characterized in that, Adjusting the operating pressure value using a preset algorithm includes: The operating pressure value is adjusted using Newton's iteration method.

11. The method as described in claim 2, characterized in that, Adjusting the hyperparameters of the neural network model based on the loss function value of the neural network model includes: Based on the loss function value of the neural network model, adjust at least one of the following: the number of hidden layers, the number of nodes in the hidden layers, the learning rate, and the number of iterations.

12. A device for predicting the red vapor pressure of gasoline, characterized in that, include: The data acquisition module is used to acquire the gasoline sample to be predicted and determine the gasoline fingerprint of the gasoline sample, wherein the gasoline fingerprint includes the molecular components of the gasoline sample; The prediction module is used to input the gasoline fingerprint into the gasoline Reid vapor pressure prediction model to obtain the gasoline Reid vapor pressure of the gasoline sample to be predicted. The gasoline Reid vapor pressure prediction model is obtained by training a neural network model using training data. The training data is obtained using an established gasoline Reid vapor pressure mechanism model. The training data includes gasoline fingerprints of preset type gasoline samples and gasoline Reid vapor pressures corresponding to the gasoline fingerprints. The gasoline Reid vapor pressure mechanism model is a calculation model obtained by simulating the experimental detection process of gasoline Reid vapor pressure and using the gas-liquid phase equilibrium principle in the experimental detection process.

13. A computer storage medium, characterized in that, The computer storage medium stores computer-executable instructions, which, when executed by a processor, implement the method for predicting the gasoline red vapor pressure according to any one of claims 1-11.

14. A terminal device for predicting the red vapor pressure of gasoline, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the method for predicting the gasoline Reid vapor pressure according to any one of claims 1-11.

Images