Methods and systems for predicting the properties of straight-run naphtha

Abstract

This invention provides a method and system for predicting the properties of straight-run naphtha, belonging to the field of petrochemical technology. The prediction method includes: obtaining the proportion of each crude oil added to the synthesis of the straight-run naphtha to be predicted, the yield of the straight-run naphtha corresponding to each crude oil, and the distillation range data; inputting the proportion of each crude oil, the yield of the straight-run naphtha corresponding to each crude oil, and the distillation range data into a pre-trained prediction model, and obtaining the first property of the straight-run naphtha to be predicted based on the distillation range segmentation model in the prediction model. Through the above technical solution, by inputting the obtained proportion of each crude oil added to the synthesis of the straight-run naphtha to be predicted, the yield of the straight-run naphtha corresponding to each crude oil, and the distillation range data into the pre-constructed process segmentation model in the prediction model, the first property of the straight-run naphtha to be predicted is determined, which shortens the naphtha property analysis cycle and improves accuracy.

Description

Technical Field

[0001] This invention relates to the field of petrochemical technology, and more specifically to a method and system for predicting the properties of straight-run naphtha. Background Technology

[0002] Straight-run naphtha is an important oil product from crude oil atmospheric and vacuum distillation units. It is a crucial feedstock for the production of ethylene and propylene via tubular furnace cracking, and for the production of benzene, toluene, and xylene via catalytic reforming. In actual production processes, near-infrared spectroscopy is typically used to analyze the properties of straight-run naphtha, but this method has a long analysis cycle and relatively low accuracy. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for predicting the properties of straight-run naphtha, in order to solve the technical problems existing in the prior art.

[0004] To achieve the above objectives, embodiments of the present invention provide a method for predicting the properties of straight-run naphtha, comprising: obtaining the proportion of each type of crude oil added to the synthesis of the straight-run naphtha to be predicted, the yield of the straight-run naphtha corresponding to each type of crude oil, and distillation range data;

[0005] The proportion of each type of crude oil, the yield of straight-run naphtha corresponding to each type of crude oil, and the distillation range data are input into a pre-trained prediction model. The first attribute of the straight-run naphtha to be predicted is obtained based on the distillation range cutting model in the prediction model.

[0006] Optionally, the distillation data includes the initial boiling point, the final boiling point, and the percentage increase in distillation temperature for each type of crude oil.

[0007] Optionally, the distillation range cutting model can be constructed according to the following formula:

[0008]

[0009] In the formula, M represents the number of crude oil types. This indicates the percentage increase in distillate volume corresponding to the distillation temperature. This indicates the yield of straight-run naphtha for each type of crude oil. This indicates the proportion of each type of crude oil.

[0010] Optionally, the distillation ratio of straight-run naphtha at a given distillation temperature for each type of crude oil can be determined according to the following formula:

[0011]

[0012] In the formula, Indicates the distillation temperature. is the coefficient, and n represents the order.

[0013] Optionally, the order can be determined based on the minimum value of the following formula:

[0014]

[0015] In the formula, SSE represents the sum of squared errors. Indicates the number of samples. This represents the true value of the distillation ratio. This represents the fitted value.

[0016] Optionally, the increase in distillation ratio P of the corresponding straight-run naphtha distilled from each type of crude oil at the corresponding distillation temperature can be determined according to the following formula. i :

[0017]

[0018] In the formula, P i This indicates the percentage increase in distillation temperature corresponding to the distillation temperature, where i is the total increase from the initial boiling point to the final boiling point.

[0019] Optionally, the prediction method further includes:

[0020] The distillation range cutting model is corrected using the rolling correction method to obtain the corrected distillation range cutting model.

[0021] Optionally, the prediction model further includes a pre-built linear prediction sub-model, and the prediction method further includes:

[0022] Obtain the type of each crude oil added to the synthetic straight-run naphtha to be predicted and the second attribute of the straight-run naphtha corresponding to each crude oil;

[0023] The type of each crude oil, the proportion of each crude oil, the second attribute of the straight-run naphtha corresponding to each crude oil, and the yield are input into the pre-constructed linear prediction sub-model to obtain the second attribute of the straight-run naphtha to be predicted.

[0024] Optionally, the first attribute includes the distillation range information of the straight-run naphtha to be predicted, and the second attribute includes at least one of the PIONA value, yield, and density of the straight-run naphtha to be predicted.

[0025] Optionally, the construction of the linear prediction sub-model includes:

[0026] Historical data for each crude oil in the blend is obtained, wherein the historical data includes the type of each crude oil, the proportion of each crude oil, and the properties and yield information of the straight-run naphtha obtained by distillation of each crude oil;

[0027] The historical detailed evaluation data of each crude oil is input into the initial linear prediction sub-model to obtain the prediction attribute information of the straight-run naphtha distilled from the mixed crude oil;

[0028] Obtain the true attribute information corresponding to the straight-run naphtha distilled from the mixed crude oil, and combine it with the predicted attribute information to correct the initial linear prediction sub-model using the rolling correction method to obtain the linear prediction sub-model.

[0029] Optionally, the linear prediction sub-model can be constructed according to the following formula:

[0030]

[0031] In the formula, n represents the number of crude oil types. This indicates the proportion of each type of crude oil. and Let Y represent a certain property and yield of straight-run naphtha obtained from the distillation of each type of crude oil, and Y represent the prediction result corresponding to that property.

[0032] Optionally, the rolling correction method includes correction according to the following formula:

[0033]

[0034] In the formula, This represents a specific attribute data output by the linear prediction sub-model or the distillation range cutting model. This indicates the actual data corresponding to this attribute. For the corrected output results, This is the correction factor.

[0035] On the other hand, the present invention also provides a system for predicting the properties of straight-run naphtha, comprising:

[0036] Memory; and

[0037] The processor is configured to execute the prediction method described above.

[0038] The above technical solution involves inputting the proportion of each type of crude oil added to the synthetic straight-run naphtha to be predicted, the yield of each type of straight-run naphtha, and the distillation process data into the pre-built prediction model's process cutting model to determine the first attribute of the straight-run naphtha to be predicted. By using the pre-trained prediction model, the analysis cycle of naphtha attributes is shortened and the accuracy is improved.

[0039] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0040] The accompanying drawings are provided to further illustrate embodiments of the present invention and form part of the specification. They are used together with the following detailed description to explain the embodiments of the present invention, but do not constitute a limitation thereof. In the drawings:

[0041] Figure 1 This is a flowchart illustrating the implementation of a method for predicting the properties of straight-run naphtha provided in an embodiment of the present invention.

[0042] Figure 2 This is a flowchart of a cutting model provided in an embodiment of the present invention;

[0043] Figure 3 This is a flowchart illustrating another method for predicting the properties of straight-run naphtha provided in this embodiment of the invention.

[0044] Figure 4 This is a flowchart of a linear prediction sub-model provided in an embodiment of the present invention;

[0045] Figure 5 This is a flowchart of a rolling correction process provided in an embodiment of the present invention;

[0046] Figure 6 This is a prediction architecture diagram of straight-run naphtha properties provided in an embodiment of the present invention;

[0047] Figure 7 (ad) is a fitting curve provided in the embodiments of the present invention when the order n is 3, 4, 5, or 6;

[0048] Figure 8 This invention provides an embodiment of the increase in the distillation ratio of naphtha for every degree Celsius increase in temperature.

[0049] Figure 9 This is a PIONA value result obtained from a linear prediction sub-model provided in an embodiment of the present invention;

[0050] Figure 10 (ad) is a curve comparing the predicted and actual values ​​of the PIONA value of No. 1 atmospheric and vacuum distillation first-grade oil provided in an embodiment of the present invention;

[0051] Figure 11 (ae) is a schematic diagram showing the comparison between the actual and predicted values ​​of the initial boiling point, 10% distillation temperature, 50% distillation temperature, 90% distillation temperature and final boiling point of No. 1 atmospheric and vacuum distillation first-stage oil provided in the embodiments of the present invention. Detailed Implementation

[0052] Ethylene is a major representative product of the petrochemical industry, occupying a dominant position among petrochemical products. Its production capacity has become an important indicator for measuring the level of a country's chemical industry internationally.

[0053] In ethylene production, feedstocks account for over 80% of the overall production cost. Different feedstocks directly determine the cost of ethylene production and also affect the yield of ethylene and by-products. Looking at the cracking feedstocks used by domestic chemical companies for ethylene production, most primarily use naphtha, accounting for over 50% of all feedstocks. Naphtha, as the main feedstock for ethylene cracking, is divided into light naphtha and heavy naphtha based on its distillation range. It is generally sourced from the company's own production and is relatively abundant. Therefore, it is essential for chemical companies to analyze the properties of straight-run naphtha to provide a basis for molecular optimization of naphtha resources, allowing for the selection of aromatic or olefinic components as appropriate.

[0054] Straight-run naphtha is an important oil product from crude oil atmospheric and vacuum distillation units. It is a crucial feedstock for the production of ethylene and propylene via tubular furnace cracking, and for the production of benzene, toluene, and xylene via catalytic reforming. The properties of straight-run naphtha include density, boiling range, sulfur content, and PIONA value. The PIONA value is a key factor in determining the flow direction of naphtha and is also an important indicator for evaluating naphtha cracking performance and determining the yield of high-value-added products from ethylene cracking. In actual production, companies typically use offline liquid chromatography (LC) to analyze the PIONA value of naphtha. However, offline analysis has a long cycle and poor timeliness, leading to delayed data acquisition and affecting naphtha quality blending. Near-infrared spectroscopy (NIRS) can achieve real-time analysis of naphtha properties, but it suffers from high equipment investment and untimely model maintenance, resulting in significant errors in the analytical results. Therefore, naphtha resource optimization urgently requires low-cost, high-precision analytical or predictive methods.

[0055] In the field of naphtha property prediction technology, both domestically and internationally, Reboucas et al. used near-infrared spectroscopy to analyze the properties of naphtha produced by the South American petrochemical company Braskem. They used a PLS calibration model to predict PINA values, density, and distillation range, with results close to actual data. Using near-infrared spectroscopy to replace chromatography, distillation, and density methods resulted in shorter measurement times, and this method has been used in the Braskem laboratory. Ding Yan et al. proposed a method for predicting naphtha properties based on online Raman spectroscopy combined with principal component particle swarm optimization (PCS). Zhou Yongjian et al. analyzed naphtha properties using online near-infrared spectroscopy, establishing near-infrared PCR and PLS prediction models constructed with different preprocessing methods. Comparing the average relative errors, they found that models preprocessed with smoothing and differential smoothing had better prediction performance. However, the applicant found that while the above solutions solved problems such as relatively long analysis times, low accuracy, or inability to extrapolate naphtha properties for petrochemical companies, they also had drawbacks: high equipment investment costs and the need for continuous model maintenance to ensure accurate predictions.

[0056] With the continuous development of machine learning technology, many scholars have attempted to apply machine learning methods to the chemical industry, modeling based on historical data to predict naphtha properties. Cheng Ming et al. established a soft-sensing model for aromatic potential using RBF neural network technology, using the density and distillation range data of heavy naphtha as network input to predict the aromatic potential content of heavy naphtha, thus solving the problem of the difficulty in measuring aromatic potential parameters. The above method can quickly and conveniently obtain prediction results, solving the problem of high cost of near-infrared spectroscopy, but its accuracy is slightly insufficient. Existing technology lacks a soft-sensing model for evaluating the properties of straight-run naphtha.

[0057] The applicant found that in actual production processes, the PIONA value analysis cycle for naphtha is too long, insufficient to meet the timeliness requirements of naphtha resource optimization; while near-infrared analysis suffers from high costs and the need for regular model maintenance and updates. To address the shortcomings of existing analytical techniques, this invention proposes a method for predicting the properties of straight-run naphtha. This method rapidly and efficiently predicts the properties of straight-run naphtha by analyzing the type and proportion of crude oil processed in atmospheric and vacuum distillation, providing auxiliary decision-making support for the optimal allocation of naphtha resources.

[0058] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the scope of the present invention.

[0059] See Figure 1 The diagram shown is a flowchart of an implementation method for predicting the properties of straight-run naphtha according to an embodiment of the present invention, including the following execution steps:

[0060] Step 100: Obtain the proportion of each crude oil added to the synthesis of the straight-run naphtha to be predicted, the yield of the straight-run naphtha corresponding to each crude oil, and the distillation range data.

[0061] In some embodiments, the distillation data includes the initial boiling point, the final boiling point, and the percentage increase in distillation temperature for each type of crude oil.

[0062] Step 101: Input the proportion of each type of crude oil, the yield of straight-run naphtha corresponding to each type of crude oil, and the distillation range data into the pre-trained prediction model, and obtain the first attribute of the straight-run naphtha to be predicted based on the distillation range cutting model in the prediction model.

[0063] In some implementations, the first attribute includes the distillation range information of the straight-run naphtha to be predicted.

[0064] In some implementations, the cutting model is constructed according to the following formula:

[0065]

[0066] In the formula, M represents the number of crude oil types. This indicates the percentage increase in distillate volume corresponding to the distillation temperature. This indicates the yield of straight-run naphtha for each type of crude oil. This indicates the proportion of each type of crude oil.

[0067] In some implementations, the distillation ratio of straight-run naphtha at a given distillation temperature for each type of crude oil is determined according to the following formula:

[0068]

[0069] In the formula, Indicates the distillation temperature. is the coefficient, and n represents the order.

[0070] In some implementations, the order is determined based on the minimum value of the following formula:

[0071]

[0072] In the formula, SSE represents the sum of squared errors. Indicates the number of samples. This represents the true value of the distillation ratio. This represents the fitted value.

[0073] In some implementations, the increase in distillation ratio P of the corresponding straight-run naphtha distilled from each type of crude oil at the corresponding distillation temperature is determined according to the following formula. i :

[0074]

[0075] In the formula, P i This indicates the percentage increase in distillation temperature corresponding to the distillation temperature, where i is the total increase from the initial boiling point to the final boiling point.

[0076] In some embodiments, the prediction method further includes: after the distillation range cutting model is constructed, the distillation range cutting model is corrected using a rolling correction method to obtain a corrected distillation range cutting model.

[0077] The rolling correction method includes correction according to the following formula:

[0078]

[0079] In the formula, This represents a specific attribute data output by the linear prediction sub-model or the distillation range cutting model. This indicates the actual data corresponding to this attribute. For the corrected output results, This is the correction factor.

[0080] In some implementations, see Figure 2 The diagram shows a flow chart of a cutting model provided by an embodiment of the present invention. The linear prediction sub-model has defects in predicting distillation range attributes, mainly in the initial boiling point and final boiling point. This is because the initial and final boiling points of the mixed oil are cut based on the initial and final boiling points of naphtha distilled from various crude oils before mixing. Cutting begins when the minimum initial boiling point temperature of the naphtha distilled from the various crude oils before mixing is reached; cutting stops when the maximum final boiling point temperature of the naphtha is reached. In other words, the initial boiling point of the naphtha produced by the atmospheric and vacuum distillation unit is the minimum initial boiling point of the naphtha distilled from all the feed crude oils of the unit that day, and the final boiling point is the maximum final boiling point of the naphtha distilled from all the feed crude oils of the unit that day. To address the problems in distillation range prediction, a naphtha distillation range cutting model for each crude oil is built based on naphtha distillation range data. The distillation range of naphtha distilled from a certain crude oil, including: initial boiling point, 10% distillation temperature, 30% distillation temperature, 50% distillation temperature… and final boiling point, is determined using a polynomial curve fitting model (…). According to the appropriate Choose the order (i.e., n) in the curve fitting model, according to the formula Determine the increased distillation percentage at the corresponding boiling point, and combine this with the crude oil type M and naphtha yield. and oil processing ratio The increase in distillation percentage (BP) of naphtha at each boiling point in an atmospheric and vacuum distillation unit is determined using the following formula. i : .

[0081] In some implementations, it is assumed that the naphtha distilled from a certain crude oil has the distillation range shown in Table 1:

[0082] Table 1. Distillation range of naphtha distilled from a certain crude oil.

[0083]

[0084] The distillation range data for each type of straight-run naphtha analysis includes seven data points from the initial boiling point to the final boiling point. The distillation percentage at the initial boiling point is 0%, and the distillation percentage at the final boiling point is 100%. To improve the accuracy of the distillation range cutting model, the temperature at which the blend achieves a certain distillation percentage is calculated by summing the distillation percentage increase for each naphtha type by one degree Celsius. To calculate the distillation percentage increase by one degree Celsius, the known distillation range data for naphtha needs to be fitted. Common fitting methods include linear regression and polynomial regression. Considering that the distillation range attribute is non-linear and the number of data points is limited, this invention uses the polynomial curve fitting method. This method, in principle, determines the expansion coefficients through least squares, and can simulate non-linearly separable data, making it more flexible than the linear regression method. A polynomial is defined as... ,in, , Let W be the coefficient of the polynomial, and n be the order of the polynomial. for The nonlinear function also represents the proportion of naphtha distilled from this type of crude oil at a given distillation temperature.

[0085] Once the objective function for fitting is determined, the fitting effect is evaluated by the sum of squared errors (SSE) to determine the order of the polynomial. The SSE calculation method is shown in the following formula, and it is desirable for the SSE value to be as small as possible.

[0086]

[0087] Where m represents the number of samples. Represents the actual value. This represents the fitted value. Calculate the SSE after fitting polynomials of different orders, considering overfitting and underfitting issues, and select the appropriate model order corresponding to the appropriate SSE.

[0088] The method for calculating the increase in the distillation ratio of naphtha for every 1°C increase is as follows: As shown, N is the total elevation from the initial boiling point to the final boiling point. This indicates the percentage increase in distillate at the corresponding distillation temperature.

[0089] For the naphtha distillation range data of each type of crude oil, a range cut-off model is established according to the above method. When predicting the range data for a certain day, the calculation method is as shown in the following formula, assuming there are M types of crude oil, based on the naphtha yield of each type of crude oil. and the increase in distillation ratio at the corresponding distillation temperature And the processing ratio of each type of crude oil. The increase in the distillation rate of naphtha at each distillation temperature in the atmospheric and vacuum distillation unit was calculated. Then, the increased distillation ratios are accumulated to obtain the correspondence between distillation temperature and distillation ratio. The predicted boiling point can be obtained based on the corresponding distillation ratio.

[0090]

[0091] This completes the construction of the distillation range cutting model.

[0092] In some implementations, see Figure 3 The diagram shows an implementation flowchart of another method for predicting the properties of straight-run naphtha provided in this embodiment of the invention. The prediction model further includes a pre-constructed linear prediction sub-model, and the prediction method specifically includes the following execution steps:

[0093] S300: Obtain the type of each crude oil added to the synthetic straight-run naphtha to be predicted and the second property of the straight-run naphtha corresponding to each crude oil.

[0094] S301: Input the type of each crude oil, the proportion of each crude oil, and the second attribute of the straight-run naphtha corresponding to each crude oil into the pre-constructed prediction sub-model to obtain the second attribute of the straight-run naphtha to be predicted.

[0095] It should be noted that the second attribute of the straight-run naphtha corresponding to each type of crude oil is the same as the second attribute of the straight-run naphtha to be predicted, which is at least one of the following: PIONA value, yield, and density.

[0096] In some implementations, the linear prediction sub-model is constructed before performing step S301, specifically through the following steps:

[0097] S1: Obtain historical data for each crude oil in the blend, wherein the historical data includes the type of each crude oil, the proportion of each crude oil, and the properties and yield information of the straight-run naphtha obtained by distillation of each crude oil;

[0098] S2: Input the historical detailed evaluation data of each crude oil into the initial linear prediction sub-model to obtain the prediction attribute information of the straight-run naphtha distilled from the mixed crude oil;

[0099] S3: Obtain the true attribute information corresponding to the straight-run naphtha distilled from the mixed crude oil, and combine it with the predicted attribute information to correct the initial linear prediction sub-model using the rolling correction method to obtain the linear prediction sub-model.

[0100] In some implementations, the rolling correction method includes correction according to the following formula:

[0101]

[0102] In the formula, This represents a specific attribute data output by the linear prediction sub-model or the distillation range cutting model. This indicates the actual data corresponding to this attribute. For the corrected output results, This is the correction factor.

[0103] In some implementations, the linear prediction sub-model in step S201 can be constructed according to the following formula:

[0104]

[0105] In the formula, n represents the number of crude oil types. This indicates the proportion of each type of crude oil. and Let Y represent a certain property and yield of straight-run naphtha obtained from the distillation of each type of crude oil, and Y represent the prediction result corresponding to that property.

[0106] In some implementations, see Figure 4 The diagram shows a flowchart of a linear prediction sub-model provided in an embodiment of the present invention. The proportion of each crude oil in the mixed crude oil, the number of crude oil types, and the raw material properties (n-alkanes, isoalkanes, cycloalkanes, aromatics, distillation range, and density) are input into the linear prediction sub-model. The linear prediction sub-model calculates the naphtha property values ​​produced by the device in a linear superposition manner based on the mixing ratio of each crude oil and the naphtha properties and yield in the crude oil detailed evaluation database. That is, the linear prediction sub-model outputs the property prediction results (including n-alkanes, isoalkanes, cycloalkanes, aromatics, distillation range, and density) of the naphtha distilled from the mixed crude oil.

[0107] In some implementations, the construction of the distillation range cutting model includes the following steps:

[0108] S0: Obtain the distillation ratio of each type of crude oil distilled at a given distillation temperature, as well as the increase in distillation ratio of each type of straight-run naphtha at the corresponding distillation temperature.

[0109] In some implementations, the distillation ratio of straight-run naphtha at a given distillation temperature for each type of crude oil can be determined according to the following formula:

[0110]

[0111] In the formula, Indicates the distillation temperature. is the coefficient, and n represents the order.

[0112] In some implementations, the order can be determined based on the minimum value of the following formula:

[0113]

[0114] In the formula, SSE represents the sum of squared errors. Indicates the number of samples. This represents the true value of the distillation ratio. This represents the fitted value.

[0115] In some implementations, the increase in distillation ratio P of the corresponding straight-run naphtha distilled from each type of crude oil at the corresponding distillation temperature can be determined according to the following formula. i :

[0116]

[0117] In the formula, P i This indicates the percentage increase in distillation temperature corresponding to the distillation temperature, where i is the total increase from the initial boiling point to the final boiling point.

[0118] S1: Based on the distillation ratio of each type of crude oil at a given distillation temperature, the increase in distillation ratio of each type of straight-run naphtha at the corresponding distillation temperature, the yield of each type of crude oil, and the proportion of each type of crude oil, the distillation range cutting model is constructed.

[0119] In some implementations, the distillation range cutting model can be constructed according to the following formula:

[0120]

[0121] In the formula, M represents the number of crude oil types. This indicates the percentage increase in distillate volume corresponding to the distillation temperature. This indicates the yield of straight-run naphtha for each type of crude oil. This indicates the proportion of each type of crude oil.

[0122] In some implementations, after the distillation range cutting model is constructed, the prediction method further includes:

[0123] The distillation range cutting model is corrected using the rolling correction method to obtain the corrected distillation range cutting model.

[0124] In some implementations, the rolling correction method includes correction according to the following formula:

[0125]

[0126] In the formula, This represents a specific attribute data output by the linear prediction sub-model or the distillation range cutting model. This indicates the actual data for this attribute. For the corrected output results, This is the correction factor.

[0127] In some implementations, parameters Figure 5 The diagram shows a rolling correction flowchart provided in an embodiment of the present invention. By comparing the naphtha PIONA values ​​obtained from offline analysis with the predicted values, some properties show significant deviations. By combining the types and proportions of crude oil processed in the atmospheric and vacuum distillation unit, rolling correction establishes a correlation between the predicted results and the offline analysis data, reducing the error in PIONA values ​​and distillation range predictions, thus achieving the "rolling" effect. In the established linear prediction model and distillation range cutting model for naphtha properties, the rolling correction method is used to improve the model's prediction accuracy. The correction method is shown in the following formula:

[0128]

[0129] in, This represents a certain attribute data output by a linear prediction model or a distillation range cutting model. This indicates the latest data for this attribute in the LIMS system. For the corrected output results, This is the correction factor.

[0130] The results predicted by the linear prediction model or the distillation range cut-off model are compared with the latest data in the LIMS system to calculate the deviation. This deviation is multiplied by a correction factor, and then added to the result predicted by the linear prediction model or the cut-off model to obtain the corrected output value. The correction factor ranges from 0 to 1. Theoretically, considering the changes in the type and proportion of crude oil actually used in refining and chemical enterprises, as well as the PIONA value and distillation range attributes analyzed by the LIMS system, the smaller the change in the type and proportion of crude oil when predicting naphtha properties, the closer the correction factor is to 1; if the difference is large, the correction factor is set to 0. Alternatively, users can manually set the correction factor based on the actual crude oil processing conditions, using the results obtained from the linear prediction model and the distillation range cut-off model without correction.

[0131] In some implementations, parameters Figure 6 The diagram shown illustrates a predictive architecture for straight-run naphtha properties provided in this embodiment of the invention. The mixing ratio, naphtha properties, and yield of each crude oil are input into a pre-built linear prediction sub-model. The distillation range data of the corresponding straight-run naphtha obtained after distillation of each crude oil are input into a pre-built cutting model. This yields the property prediction results for the straight-run naphtha obtained after distillation of the mixed crude oils. Based on the established linear prediction sub-model and distillation range cutting model for naphtha properties, a rolling correction method is used, combined with the system's original data (i.e., real data), to correct the linear prediction sub-model and cutting model, thereby improving the model's prediction accuracy and obtaining more accurate naphtha property prediction results.

[0132] In some embodiments, to further highlight the advantages of the present invention, some technical contents of specific embodiments of the present invention will be illustrated below:

[0133] First, the distillation range data of naphtha distilled from crude oil 1 was selected to build a distillation range cutting model for the naphtha. The distillation range data of naphtha distilled from crude oil 1 is shown in Table 2.

[0134] Table 2. Distillation range of naphtha from crude oil 1

[0135]

[0136] Using the polynomial fitting method, by selecting the polynomial order, the distillation range and distillation ratio of naphtha obtained from crude oil distillation were fitted respectively, and fitting curves of different orders were obtained, as shown in the figure. Figure 7 As shown in (ad). Figure 7 (a), 7(b), 7(c), and 7(d) represent the fitted curves corresponding to orders n of 3, 4, 5, and 6, respectively. Once the objective function for fitting is determined, the polynomial order is determined by evaluating the fitting effect using the sum of squared errors (SSE). The relationship between the polynomial order and SSE is shown in Table 3.

[0137] Table 3 Relationship between order and SSE variation

[0138]

[0139] pass Figure 7 Based on the comprehensive evaluation in Table 3, to prevent overfitting and underfitting of the curves, the polynomial order n was determined to be 5. The correspondence between distillation temperature and distillation ratio after polynomial fitting is shown in the following formula. This indicates that the naphtha distilled from crude oil 1 at a given distillation temperature The corresponding distillation ratio at that time.

[0140]

[0141] The distillation volume of naphtha from crude oil 1, from the initial boiling point to the final boiling point, was calculated using the distillation range fitting formula. Then, the increase in distillation percentage for each degree Celsius increase in temperature was calculated. The results are as follows: Figure 8 As shown, it can be seen that for every degree Celsius increase in temperature, the increased distillation ratio of the naphtha exhibits a distribution trend of first increasing and then decreasing.

[0142] Based on the types and proportions of crude oil processed by the No. 1 atmospheric and vacuum distillation unit in the daily production report of a refining enterprise for a certain period, and substituting them into the linear prediction sub-model, the resulting PIONA value is as follows: Figure 9 As shown in the figure, by comparing the actual PIONA data in the LIMS system, the predicted n-alkane content is too high, while the cycloalkanes and aromatics content is too low, and the difference from the actual values ​​is significant.

[0143] The results predicted by the linear prediction model or the distillation range cut-off model are compared with the latest corresponding analysis data in the LIMS system. The deviation is calculated, multiplied by a correction factor, and then added to the result predicted by the linear prediction model or the cut-off model to obtain the corrected output value. Considering that the types and proportions of crude oil in actual feed are relatively complex and the relationship with the correction factor is difficult to determine, this invention simplifies the setting of the correction factor when validating the model, dividing it into the following three cases: the type and proportion of crude oil do not change or change only slightly, the correction factor is 1; the type and proportion of crude oil change significantly, the correction factor is 0.8; and the type of crude oil is different, the correction factor is 0. By comparing the model prediction results with the actual data, it is found that the model meets the accuracy requirements. The prediction results are analyzed below.

[0144] This invention takes the No. 1 atmospheric and vacuum distillation unit as an example, selecting measured data of PIONA and distillation range of the No. 1 atmospheric and vacuum distillation unit's atmospheric top primary oil from July 1st to October 18th of a certain year in the LIMS system, and comparing them with the results output by the calibrated prediction model to verify the model's prediction effect. The comparison curve of predicted and actual PIONA values ​​for the No. 1 atmospheric and vacuum distillation unit's atmospheric top primary oil is shown below. Figure 10 As shown in (ad). Figure 10 Figures (a), 10(b), 10(c), and 10(d) represent the comparison between the actual and predicted values ​​of n-alkanes, isoalkanes, cycloalkanes, and aromatics in the No. 1 atmospheric and vacuum distillation oil (atmospheric and vacuum distillation top-stage oil). The figures show that the trends of the corrected results are generally consistent with the measured data, with only a few slightly larger differences. The curves comparing the predicted and actual values ​​of the No. 1 atmospheric and vacuum distillation oil's distillation range are shown below. Figure 11 As shown in (ae).

[0145] Figure 11 Figures (a), (b), (c), (d), and (e) represent the comparison between the actual and predicted values ​​of the initial boiling point, 10% distillation temperature, 50% distillation temperature, 90% distillation temperature, and final boiling point of No. 1 atmospheric and vacuum distillation grade 1 oil. The figures show that the predicted trend of the distillation range is basically consistent with the actual data. To verify the accuracy of the model, three evaluation indicators were used: mean relative error (MRE), mean absolute error (MAE), and mean squared error (MSE). The calculation methods are shown in the following formulas.

[0146]

[0147]

[0148]

[0149] Where N is the number of samples evaluated. This represents the difference between the actual value and the fitted value, i.e., the residual. The MRE, MAE, and MSE values ​​between the predicted and actual values ​​of the PIONA value and distillation range are calculated, and the results are shown in Table 4.

[0150] Table 4 Evaluation metrics for the prediction model

[0151]

[0152] As shown in Table 4, except for cycloalkanes, the MRE of other attributes is within 5%, and the MAE and MSE are also relatively small. Considering the attribute requirements for subsequent naphtha configuration optimization modeling, the PIONA value is a key indicator affecting the destination of naphtha, and the indicators for n-alkanes and isoalkanes are more important. Therefore, the relative error of the predicted n-alkanes and isoalkanes needs to be less than 5%, while the relative error for aromatics and cycloalkanes, which are slightly lower, needs to be less than 10%. Thus, the model meets the accuracy requirements. In addition, the model is more accurate in predicting the distillation range, and the prediction results of the model are related to the frequency of naphtha attribute analysis in the LIMS system. During calibration, the latest time point data in the LIMS system is used as the basis, while the PIONA analysis cycle for naphtha from atmospheric and vacuum distillation units in the LIMS system is relatively long, and the frequency of distillation range analysis is high.

[0153] On the other hand, embodiments of the present invention also provide a system for predicting the properties of straight-run naphtha, comprising:

[0154] Memory; and

[0155] The processor is configured to execute the prediction method described in any of the above embodiments.

[0156] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0157] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0158] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0159] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0160] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0161] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0162] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0163] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0164] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for predicting the properties of straight-run naphtha, characterized in that, include: Obtain the proportion of each crude oil added to the synthetic straight-run naphtha to be predicted, the yield of each crude oil corresponding to the straight-run naphtha, and the distillation range data; The proportion of each type of crude oil, the yield of straight-run naphtha corresponding to each type of crude oil, and the distillation range data are input into a pre-trained prediction model. Based on the distillation range cutting model in the prediction model, the first attribute of the straight-run naphtha to be predicted is obtained. The distillation range cutting model is constructed based on the following formula: In the formula, This indicates the percentage increase in naphtha yield per distillation temperature, where M represents the number of crude oil types. This indicates the percentage increase in distillate volume corresponding to the distillation temperature. This indicates the yield of straight-run naphtha for each type of crude oil. This indicates the proportion of each type of crude oil, where N is the total increase in boiling point from the initial boiling point to the final boiling point; The following formula can be used to determine the distillation ratio of straight-run naphtha from each type of crude oil at a given distillation temperature: In the formula, Indicates the distillation temperature. Here, is the coefficient, and n represents the order; The order is determined by the minimum value of the following formula: In the formula, SSE represents the sum of squared errors. Indicates the number of samples. This represents the true value of the distillation ratio. Indicates the fitted value; The following formula is used to determine the percentage increase in distillation temperature (P) of the corresponding straight-run naphtha distilled from each type of crude oil. i : In the formula, P i This indicates the percentage increase in distillate temperature corresponding to the distillation temperature, where N is the total increase from the initial boiling point to the final boiling point.

2. The prediction method according to claim 1, characterized in that, The distillation data includes the initial boiling point, the final boiling point, and the percentage increase in distillation temperature for each type of crude oil.

3. The prediction method according to claim 1, characterized in that, The prediction method further includes: The distillation range cutting model is corrected using the rolling correction method to obtain the corrected distillation range cutting model.

4. The prediction method according to claim 1, characterized in that, The prediction model also includes a pre-built linear prediction sub-model, and the prediction method further includes: Obtain the type of each crude oil added to the synthetic straight-run naphtha to be predicted and the second attribute of the straight-run naphtha corresponding to each crude oil; The type of each crude oil, the proportion of each crude oil, and the second attribute of the straight-run naphtha corresponding to each crude oil are input into the pre-constructed linear prediction sub-model to obtain the second attribute of the straight-run naphtha to be predicted.

5. The prediction method according to claim 4, characterized in that, The first attribute includes the distillation range information of the straight-run naphtha to be predicted, and the second attribute includes at least one of the PIONA value, yield, and density of the straight-run naphtha to be predicted.

6. The prediction method according to claim 4, characterized in that, The construction of the linear prediction sub-model includes: Historical data for each crude oil in the blend is obtained, wherein the historical data includes the type of each crude oil, the proportion of each crude oil, and the properties and yield information of the straight-run naphtha obtained by distillation of each crude oil; The historical detailed evaluation data of each crude oil is input into the initial linear prediction sub-model to obtain the prediction attribute information of the straight-run naphtha distilled from the mixed crude oil; Obtain the true attribute information corresponding to the straight-run naphtha distilled from the mixed crude oil, and combine it with the predicted attribute information to correct the initial linear prediction sub-model using the rolling correction method to obtain the linear prediction sub-model.

7. The prediction method according to claim 4, characterized in that, The linear prediction sub-model is constructed according to the following formula: In the formula, n represents the number of crude oil types. This indicates the proportion of each type of crude oil. and Let Y represent a certain property and yield of straight-run naphtha obtained from the distillation of each type of crude oil, and Y represent the prediction result corresponding to that property.

8. The prediction method according to claim 3 or 6, characterized in that, The rolling correction method includes correction according to the following formula: In the formula, This represents a specific attribute data output by the linear prediction sub-model or the distillation range cutting model. This indicates the actual data corresponding to this attribute. For the corrected output results, This is the correction factor.

9. A predictive system for the properties of straight-run naphtha, characterized in that, include: Memory; as well as The processor is configured to perform the prediction method as described in any one of claims 1-8.

Images