CN110021384A - A method of for predicting octane number - Google Patents

Abstract

The embodiment of the present invention provides a kind of method for predicting octane number, belongs to petrochemical industry.The described method includes: calculating active nucleus conversion ratio according to each component in the gasoline and active nucleus conversion ratio computation model；And according to the active nucleus conversion ratio and octane number computation model, calculate the octane number.Combustion chemistry model of the program by research hydro carbons in the cylinder, decomposition combustion process, propose the guess that active nucleus conversion ratio in system determines octane number size, and carry out calculated octane number based on this, so as to consider that non-linear blending effect may be contributed or be lost caused by octane number during each blending component blending in gasoline, so that the prediction of octane number is more accurate.

Description

A method for predicting gasoline octane number

technical field

The invention relates to the petrochemical field, in particular to a method for predicting the octane number of gasoline.

Background technique

Gasoline is prepared by blending various components. The inventors of the present application found in the process of realizing the present invention that no matter whether the blending component here refers to a pure hydrocarbon compound or a certain component oil, the octane number during the blending process There are obvious nonlinear laws. It can be said that the octane number of gasoline is not only related to the octane number of each blending component in gasoline, but also to the blending characteristics of each component in the blending process. In recent years, in the process of upgrading oil products, more emphasis has been placed on the addition ratio of high-octane components, while relatively neglecting the contribution or loss of non-linear blending effects to gasoline octane number during the blending process. The octane number prediction model based on the detailed composition aims to solve this problem. While understanding the gasoline octane number at the molecular level, it can achieve the goal of high-precision prediction of the octane number. The core of this model is the establishment of the mathematical expression relationship between gasoline composition and octane number.

At present, some domestic and foreign research institutions have given a small amount of conjectures on the mathematical relationship between octane number and gasoline composition. However, these models are mostly deduced from assumptions made by experimental laws, and lack of in-depth understanding and theoretical research on the octane number blending process. , limited by the basis of the experimental method, the obtained model also has shortcomings in terms of prediction accuracy and scope of application.

SUMMARY OF THE INVENTION

The purpose of the embodiments of the present invention is to provide a method for predicting gasoline octane number, which can improve the prediction accuracy of gasoline octane number.

In order to achieve the above object, an embodiment of the present invention provides a method for predicting the octane number of gasoline, the method comprising: calculating the active core conversion rate according to the calculation model of each component in the gasoline and the active core conversion rate; and according to the calculation model of the active core conversion rate; The active core conversion rate and the octane number calculation model are used to calculate the gasoline octane number.

Optionally, before calculating the gasoline octane number according to the active core conversion rate and the octane number calculation model, the method further includes: correcting according to the interaction relationship of each component in the gasoline and the active conversion rate. model, corrected for the active nuclear conversion rate.

In another aspect, the present invention provides a machine-readable storage medium, where instructions are stored on the machine-readable storage medium, the instructions are used to cause a machine to execute the above-mentioned method for predicting gasoline octane number of the present application.

The octane number is an indicator of the anti-knock ability of the reaction gasoline against the amount of n-heptane and isooctane. The inventor of the present case recognizes that: 1) octane number is a relative concept; 2) octane number is not only related to the composition of gasoline, but also to the reaction chemistry in the combustion process, which is the reason for the nonlinearity of octane number in the blending process . The invention provides a method for establishing a mathematical relationship expression between gasoline composition and octane number: by means of studying the combustion chemical model of hydrocarbons in a cylinder and decomposing the combustion process, it is proposed that the conversion rate of active cores in the system determines the octane number. Conjecture, and calculate the octane number based on this, so that the contribution or loss of the non-linear blending effect in the blending process of each blending component in gasoline may be considered to the gasoline octane number, so that the gasoline octane number predictions are more accurate.

Additional features and advantages of embodiments of the present invention will be described in detail in the detailed description section that follows.

Description of drawings

The accompanying drawings are used to provide a further understanding of the embodiments of the present invention, and constitute a part of the specification, and are used to explain the embodiments of the present invention together with the following specific embodiments, but do not constitute limitations to the embodiments of the present invention. In the attached image:

1 is a flowchart of a method for predicting gasoline octane number provided by an embodiment of the present invention;

2 is a flowchart of a method for predicting gasoline octane number provided by another embodiment of the present invention;

Fig. 3 is the effect diagram of carrying out octane number prediction according to the method for predicting gasoline octane number provided according to the first embodiment of the present invention;

Fig. 4 is the effect diagram of carrying out octane number prediction according to the method for predicting gasoline octane number provided according to the second embodiment of the present invention;

Fig. 5 is the effect diagram of carrying out octane number prediction according to the method for predicting gasoline octane number provided by the third embodiment of the present invention;

6A and 6B are effect diagrams of octane number prediction performed according to the method for predicting gasoline octane number provided by the fourth embodiment of the present invention;

7A and 7B are effect diagrams of octane number prediction performed according to the method for predicting gasoline octane number provided by the fifth embodiment of the present invention;

8 is an effect diagram of performing octane number prediction according to the method for predicting gasoline octane number provided by the sixth embodiment of the present invention; and

FIG. 9 is an effect diagram of octane number prediction according to the method for predicting gasoline octane number provided by the seventh embodiment of the present invention.

Detailed ways

The specific implementations of the embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific implementation manners described herein are only used to illustrate and explain the embodiments of the present invention, and are not used to limit the embodiments of the present invention.

FIG. 1 is a flowchart of a method for predicting gasoline octane number provided by an embodiment of the present invention. As shown in FIG. 1 , an embodiment of the present invention provides a method for predicting the octane number of gasoline, the method comprising: calculating an active core conversion rate according to a calculation model of each component in the gasoline and an active core conversion rate; and According to the active nuclear conversion rate and the octane number calculation model, the gasoline octane number is calculated.

By studying the combustion chemical model of hydrocarbons in the cylinder, the scheme decomposes the combustion process, and proposes a conjecture that the active nuclear conversion rate in the system determines the octane number, and based on this, the octane number is calculated, so that gasoline can be considered. The contribution or loss of the non-linear blending effect to gasoline octane number during the blending process of each blending component in , makes the prediction of gasoline octane number more accurate. This scheme simplifies the combustion reaction process of gasoline before low temperature flame by means of the combustion mechanism, introduces the assumption that the active nuclear conversion rate affects the octane number, and divides the octane number model into the calculation of the active nuclear conversion rate and the modification of the active nuclear conversion rate. , and the relationship between the octane number and the conversion rate of active nuclear three steps to model each step, and finally combined into the mathematical form of the octane number mechanism model based on the detailed hydrocarbon composition.

In addition to considering the influence of each component on the conversion rate of active nuclei in the system, the components will also interact during the combustion process, adding new active nuclei or inert nuclei to generate pathways, and then affecting the conversion rate of active nuclei in the system. make an impact. FIG. 2 is a flowchart of a method for predicting gasoline octane number provided by another embodiment of the present invention. Preferably, as shown in FIG. 2, before calculating the gasoline octane number according to the active core conversion rate and the octane number calculation model, the method further includes: according to the interaction relationship of each component in the gasoline and the active transformation rate correction model, the active nuclear transformation rate is corrected.

The three models involved in the above technical solutions are introduced separately as follows:

Model 1: Calculation model of active nuclear conversion rate

Specifically, the active nuclear conversion rate calculation model can be selected from one of the following to calculate the active nuclear conversion rate Q _ac produced by unit mole of components in the system:

Model 1A:

Among them, Q _ac is the conversion rate of active nuclei produced by unit mole of component, [ _ni ] represents the content of active nuclei produced by component i, _ni is the mole fraction of component i, υ _i is the volume fraction of component i, K _i is the conversion rate of pure component i to form active nuclei at the end of the reaction before the low temperature flame, β _i is the blending factor of i component, where ρ _i is the relative density of the _i component, and Mi is the relative molecular weight of the i component. The value of β _i ultimately needs to be regressed through experimental data, but through simple derivation, it can be seen that its value is related to the ratio of the density and molecular weight of the i component. From the definition of octane number, it can be known that the β _i value of isooctane and n-heptane is 1, a reference function can be established from this, and the initial value of β _i for each component can be obtained.

Model 1B:

where n _i is the mole fraction of free radicals generated by component i in the pre-flame reaction stage, n _i is the molar fraction of component i, θ _i is the reaction rate of free radicals generated by component i in the pre-low temperature flame reaction stage, q _i is the reaction rate at which the i component radicals further generate active nuclei. The model is based on the assumption that the ratio of active nuclei to free radicals (ie, the ability of free radicals to become active nuclei) determines the octane number.

Model 1C:

Among them, the meanings of the parameters in this model 1C are the same as those in the above-mentioned model 1B. This model is a model formed by adding the assumption of competitive oxidation of components on the basis of model 1B and referring to the adsorption mechanism.

Model 2: Active Conversion Rate Correction Model

In addition to considering the influence of each component on the conversion rate of active nuclei in the system, the components will also interact during the combustion process, adding new active nuclei or inert nuclei to generate pathways, and then affecting the conversion rate of active nuclei in the system. make an impact. The steady-state equation can be established through different assumptions about the mechanism, and the functional relationship between the newly added active nucleus and its composition can be obtained. Two model forms, 2A and 2B, are given here, which can calculate the number of newly added active nuclei and add them to the above calculation model of active nuclei conversion rate to complete the correction of the active nuclei conversion rate Q _ac .

The active conversion rate correction model can be selected from one of the following to correct for the active nuclear conversion rate Q _ac :

Model 2A:

Model 2A is derived from the steady state equation established by reaction mechanism A, such as 2A', where k _i , are the reaction rates of the reaction mechanism A, respectively, n _i , n _j are the mole fractions of the i component and the j component, respectively, is the newly added active nuclear transformation rate of the i component.

When examining the combustion reaction of a certain component, the model assumes that other components are the catalytic factors of the reaction and affect the reaction process of this component. Among them, A, B, C represent the blending components, [A] represents the active core produced by the A component through this pathway, Represents the hydroxyl radical generated by the active nucleus through a series of branched chain reactions. In this process, once the active nucleus is formed, a branch chain reaction will occur rapidly to generate a large number of hydroxyl radicals, and the latter rapidly exothermic combustion produces detonation, and how each blending component generates an active nucleus is the reaction before the low temperature flame. The rate-determining step, so the active nuclear conversion rate is an important indicator that affects the knocking.

Model 2B:

Model 2B is derived from the steady state equation established by the reaction mechanism B. The steady state equation is shown in the following 2B', t _ij is the interaction parameter of the i component and j component, n _i , n _j are i respectively the mole fractions of component and j component, is the newly added active nuclear conversion rate of the system.

The model considers that when examining the combustion reaction of a certain component, the interaction between the two components in the system promotes the generation of active nuclei or inert nuclei in the system and affects the proportion of total active nuclei. Among them, A, B, C represent the blending components, [M] represents the newly generated active nucleus of the pathway, Represents the hydroxyl radical generated by the active nucleus through a series of branched chain reactions.

Model 3: Octane Number Calculation Model

According to the analysis of the detonation principle, it is believed that the conversion rate of active nuclei is positively correlated with the detonation intensity. According to the octane number standard, the knock intensity is negatively correlated with the octane number, so the active nuclear conversion rate is negatively correlated with the octane number. Based on this, a variety of model conjectures about the relationship between active nuclear conversion rate and octane number can be proposed. The present invention provides four octane number calculation models, and the octane number calculation models are selected from one of the following four types:

Model 3A (linear function): RON=aQ _ac +b

Model 3B (reciprocal function): RON=a/Q _ac +b

Model 3C (quadratic function): RON=a(Q _ac +b) ² +c

Model 3D (exponential function): RON=exp(aQ _ac +b)

where RON is octane number, Q _ac is active nuclear conversion, and a, b, and c are correction parameters. In modeling, these parameters should be adjusted to ensure the negative correlation between octane number and active nuclear conversion rate.

The above three models are summarized in the following table:

Finally, the mathematical relationship of the octane number prediction model can be a combination of these three parts of the model, and the combination is as follows:

Octane = Model 3 (Model 1 + Model 2)

When combined, Models 1 and 2 are optional parts, or they can be completely ignored. When only model 3 is considered and linear model 3A is used, the resulting octane model is the simplest linear model. When such a prediction model is used for a single component, the _octane number on the left side of the formula is the _octane number of the pure component. most of the unknown parameters to achieve the purpose of simplifying the model. In addition, when using model 2 to correct the conversion rate of active nuclei in model 1, the newly added active nuclei can be added to the molecules of model 1, or:

Convert the amount of newly added active cores to the newly added mole fraction of i component, at this time:

n _i =(1+I _i )n _i or n _i =(1+I _mix /n _i )n _i

After doing the above processing, the corrected i component mole fraction was substituted into Model 1 to calculate the active nucleus conversion rate.

The present invention can separately model the above-mentioned three models and make multiple model guesses, and use the above-mentioned three models to combine to form the final octane number prediction model. Theoretical explanation and initial value of each parameter in the model. Finally, the parameters in the octane number prediction model can be corrected by using the refined oil data, and finally a complete octane number prediction model based on the detailed composition can be obtained.

Three embodiments of establishing the mathematical expression of the octane number prediction model by the present invention are given below.

Example 1

For the three parts of the combination, model 1A, model 2A, and model 3A are selected respectively, and the octane number of pure hydrocarbons is substituted, and the intermediate parameters are simplified to obtain the octane number prediction model expression 1A-2A-3A based on the detailed composition of gasoline:

in, P represents the component that is considered to participate in the correction of the active nuclear transformation rate of Model 2, where ρ _i is the relative density of the _i component, and Mi is the relative molecular weight of the i component. ON _i is the octane number of each pure component, which is a known parameter, υ _i is the volume fraction of the i component, β _i and a are the parameters to be regressed by the model, and the initial value of β _i can be obtained from the i component Density and molecular weight were obtained.

The model expression obtained by this combination is similar to the formula obtained by Exxon Company through experiments, but the parameters of the formula obtained by this method have practical significance, and the method for obtaining the initial value of the key parameter β _i is given: According to the octane number The test method shows that the β _i of _n -heptane and isooctane is 1, and the density and molecular weight are also known _.

Using the data of 194 refined oil samples and 67 component oil samples obtained by us, the mathematical expression has good prediction accuracy for octane number. As shown in Figure 3, the abscissa is the octane number actually measured by the sample, the ordinate is the octane number calculated by the model, "*" is the model parameter training set, "+" is the model parameter test set, and the results show standard The deviation is 0.513.

Embodiment 2

For the three parts of the combination, model 1A, model 2B, and model 3A are selected respectively, and the octane number of pure hydrocarbons is substituted, and the intermediate parameters are simplified to obtain the octane number prediction model expression 1A-2B-3A based on the detailed composition of gasoline:

where, I _mix =∑ _ij t _{ij n in j ,} t _ij is the interaction parameter of the i component and the j component that needs to be considered in the correction of the active nuclear conversion rate of Model 2, which is obtained from the model regression.

Similarly, the method in the first embodiment is used to obtain the initial value of the parameter β _i , the data is used to regress the parameters, and the prediction effect of the expression is verified. As shown in Figure 4, the abscissa is the octane number actually measured by the sample, the ordinate is the octane number calculated by the model, "*" is the model parameter training set, "+" is the model parameter test set, and the results show standard The deviation is 0.488.

Embodiment 3

For the three parts of the combination, model 1A, model 2A, and model 3B are selected respectively, and the octane number of pure hydrocarbons is substituted, and the intermediate parameters are simplified to obtain the octane number prediction model expression 1A-2A-3B based on the detailed composition of gasoline:

The parameters in the formula are the same as those in the first embodiment.

Similarly, the method in the first embodiment is used to obtain the initial value of the parameter β _i , the data is used to regress the parameters, and the prediction effect of the expression is verified. The octane number calculated by the model, "*" is the model parameter training set, "+" is the model parameter test set, and the result shows that the standard deviation is 0.532.

Embodiment 4

Models 1A and 3A can be selected respectively. For pure hydrocarbons, model 1A can be expressed as:

Q _ac =K _i (1)

The left side of Model 3A can be replaced by the i-component pure hydrocarbon octane number ON _i , which becomes:

ON _i = aQ _ac + b (2)

Substituting formula (1) into formula (2), the relationship between pure hydrocarbon octane number ON _i and K _i can be established:

Substituting model 1A and (3) into model 3A can eliminate most of the unknown parameters in the model and achieve the purpose of simplifying the model. The final octane number model is:

Among them, ON _i is the octane number of each pure component as a known parameter, v _i is the volume fraction of the i component, and β _i is the blending factor of the i component, which is the parameter that the model needs to regress. By training the model with a small amount of composition and octane number fact data, and obtaining the β _i parameter, the gasoline composition can be used to predict its octane number.

For the combination of models 1A and 3A, the following two examples can be carried out to verify its effectiveness.

Example 1), 194 refined oil samples and 67 component oil sample data were used to validate the model, 20 of which were used as training sets, and the remaining data were used as validation sets. After using the data to regress the model parameters, the model has better prediction accuracy for the octane number of the sample. As shown in Figure 6A, the abscissa is the octane number actually measured by the sample, the ordinate is the octane number calculated by the model, "*" is the model parameter training set, and "+" is the model parameter test set. The standard deviation is 0.663.

Embodiment 2), the model is used for the prediction of 6 groups of reforming component oils and 7 groups of catalytic cracking component oils, the model parameters adopt the parameters obtained by regression in the above-mentioned embodiment 1), and the model has a relatively high octane number to the sample. good prediction accuracy. As shown in Figure 6B, the abscissa is the octane number actually measured by the sample, the ordinate is the octane number calculated by the model, "*" is the model parameter training set, and "+" is the model parameter test set. The standard deviation is 0.371.

Embodiment 5

Models 1C and 3A can be selected respectively, and the two are combined and mathematically calculated to simplify the intermediate parameters, and the final octane number prediction expression model is obtained as follows:

Among them, _ni is the mole fraction of free radicals generated by component i in the reaction stage before the low temperature flame, _{ni is the mole fraction of component i, θ i is} the reaction rate of the free radicals generated by component i in the reaction stage before the low temperature flame, ON _i The octane number of each pure component is used as a known parameter.

For the combination of models 1C and 3A, the following two examples can be performed to verify its effectiveness.

Example 1), 194 refined oil samples and 67 component oil sample data were used to validate the model, 40 of which were used as training sets, and the remaining data were used as validation sets. After using the data to regress the model parameters, the model has better prediction accuracy for the octane number of the sample. As shown in Figure 7A, the abscissa is the octane number actually measured by the sample, the ordinate is the octane number calculated by the model, "*" is the model parameter training set, and "+" is the model parameter test set. The standard deviation is 0.649.

Embodiment 2), the model is used for the prediction of 70 catalytic cracking component oils of a certain refinery, the model parameters adopt the parameters obtained by regression in the above-mentioned embodiment 1), and the model has better prediction accuracy to the octane number of the sample. . As shown in Figure 7B, the abscissa is the octane number actually measured by the sample, the ordinate is the octane number calculated by the model, "*" is the model parameter training set, and "+" is the model parameter test set. The standard deviation is 0.412.

Embodiment 6

Models 1B and 3A can be selected respectively, the two are combined and the intermediate parameters are simplified after mathematical calculation, and the final octane number prediction expression model is obtained as follows:

Among them, ON _i is the octane number of each pure component as a known parameter, n _i is the mole fraction of i component, θ _i is the reaction rate of the i component to generate free radicals in the reaction stage before the low temperature flame, which is the model that needs to be regressed parameter. By training the model with a small amount of composition and octane number fact data, and obtaining the _θi parameter, the octane number can be predicted from the gasoline composition.

The data of 194 refined oil samples and 67 component oil samples are used to verify that the mathematical expression has good prediction accuracy for octane number. As shown in Figure 8, the abscissa is the octane number actually measured by the sample, the ordinate is the octane number calculated by the model, "*" is the model parameter training set, "+" is the model parameter test set, and the results show standard The deviation is 0.663.

Embodiment 7

Models 1A and 3B can be selected respectively, and the two are combined and mathematically calculated to simplify the intermediate parameters, and the final octane number prediction expression model is obtained as follows:

Among them, ON _i is the octane number of each pure component as a known parameter, v _i is the volume fraction of i component, and β _i is the parameter to be regressed by the model. By training the model with a small amount of composition and octane number fact data, and obtaining the value of the parameter β _i , the gasoline composition can be used to predict its octane number.

The data of 194 refined oil samples and 67 component oil samples are used to verify that the mathematical expression has good prediction accuracy for octane number. As shown in Figure 9, the abscissa is the octane number actually measured by the sample, the ordinate is the octane number calculated by the model, "*" is the model parameter training set, "+" is the model parameter test set, and the results show standard The deviation is 0.681.

Correspondingly, an embodiment of the present invention further provides a machine-readable storage medium, where instructions are stored on the machine-readable storage medium, and the instructions are used to cause a machine to execute the above-mentioned method for predicting gasoline octane number of the present application.

The optional embodiments of the embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the embodiments of the present invention are not limited to the specific details of the above-mentioned embodiments. A variety of simple modifications are made to the technical solution of the invention, and these simple modifications all belong to the protection scope of the embodiments of the present invention.

In addition, it should be noted that each specific technical feature described in the above-mentioned specific implementation manner may be combined in any suitable manner under the circumstance that there is no contradiction. To avoid unnecessary repetition, various possible combinations are not further described in this embodiment of the present invention.

Those skilled in the art can understand that all or part of the steps in the method of the above-mentioned embodiments can be completed by instructing the relevant hardware through a program, and the program is stored in a storage medium and includes several instructions to make a single-chip microcomputer, a chip or a processor. (processor) executes all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: U disk, removable hard disk, Read-Only Memory (ROM, Read-Only Memory), Random Access Memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program codes .

In addition, various implementations of the embodiments of the present invention may also be combined arbitrarily, as long as they do not violate the ideas of the embodiments of the present invention, they should also be regarded as the contents disclosed in the embodiments of the present invention.

Claims (3)

1. A method for predicting the octane number of a gasoline, the method comprising:

calculating the active nuclear conversion rate according to each component in the gasoline and an active nuclear conversion rate calculation model; and

and calculating the octane number of the gasoline according to the active nuclear conversion rate and an octane number calculation model.

Wherein the calculation model of the active nuclear conversion rate is as follows:

model 1C:

wherein n is_iIs the molar fraction of radicals, θ, produced by the i component during the low temperature pre-flame reaction stage_iIs a low temperature pre-flame reaction stage, i component reaction rate for generating free radicals, q_iIs the reaction rate of the i component free radicals to further generate active cores,

the octane calculation model is selected from one of:

model 3A: RON ═ aQ_ac+b

Wherein RON is the octane number, Q_acIs the active nucleus conversion, and a, b, c are correction parameters.

2. The method of claim 1, wherein model 1C and model 3A constitute the following octane number predictive expression models:

wherein n is_iIs the molar fraction of radicals, n, produced by the i component in the low-temperature pre-flame reaction stage_iIs the molar fraction of the i component, θ_iIs the reaction rate, ON, of the formation of free radicals of the component i in the low-temperature pre-flame reaction stage_iThe octane number of each pure component is a known parameter.

3. A machine-readable storage medium having stored thereon instructions for causing a machine to perform the method of any one of claims 1-2 above.