JP2024081328A - Information processing device, prediction method, and program - Google Patents

Abstract

[Problem] To predict the amount of naphthenobenzene in kerosene produced by blending multiple base oils. [Solution] An information processing device 10 includes a memory unit 14 that stores model data 24 relating to a correlation model that shows the correlation between multiple parameters indicating the properties of each of multiple kerosenes, the multiple parameters including at least density, distillation properties, and aromatic content, and the amount of naphthenobenzene contained in the multiple kerosenes, an acquisition unit 12 that acquires the numerical values of the multiple parameters of a specific kerosene, and a calculation unit 16 that predicts the amount of naphthenobenzene contained in the specific kerosene using the model data 24 and the numerical values of the multiple parameters of the specific kerosene. [Selected Figure] Figure 1

Description

The present disclosure relates to an information processing device, a prediction method, and a program.

In recent years, the demand for petroleum products has shifted to lighter oils, and there is a demand to increase the production of light oils such as kerosene by utilizing heavy oils and residual oils. Cracked base kerosene obtained by cracking heavy oils and residual oils tends to contain a high proportion of naphthenobenzenes. It is known that an increase in the amount of naphthenobenzene contained in kerosene can hinder the stable long-term operation of heating equipment such as kerosene stoves. When using cracked base kerosene, it is said that the amount of naphthenobenzene contained in the kerosene needs to be properly managed (see, for example, Patent Document 1).

JP 2015-183030 A

Because analyzing naphthenobenzene content is complicated and time-consuming, it is difficult to sequentially analyze the amount of naphthenobenzene contained in kerosene at kerosene manufacturing sites.

One exemplary objective of an embodiment of the present disclosure is to provide a technology for easily predicting the amount of naphthenobenzene contained in kerosene.

An information processing device according to one aspect of the present disclosure includes a memory unit that stores model data relating to a correlation model that indicates a correlation between a plurality of parameters indicating the properties of each of a plurality of kerosenes, the plurality of parameters including at least density, distillation properties, and aromatic content, and the amount of naphthenobenzene contained in each of the plurality of kerosenes; an acquisition unit that acquires the numerical values of the plurality of parameters of a specific kerosene; and a calculation unit that predicts the amount of naphthenobenzene contained in the specific kerosene using the model data and the numerical values of the plurality of parameters of the specific kerosene.

Another aspect of the present disclosure is a prediction method. This method includes: acquiring model data relating to a correlation model that indicates a correlation between a plurality of parameters indicating properties of each of a plurality of kerosenes, the plurality of parameters including at least density, distillation properties, and aromatic content, and the amount of naphthenobenzene contained in each of the plurality of kerosenes; acquiring numerical values of the plurality of parameters for a specific kerosene; and predicting the amount of naphthenobenzene contained in the specific kerosene using the model data and the numerical values of the plurality of parameters for the specific kerosene.

Yet another aspect of the present disclosure is a program. This program has a function of acquiring model data relating to a correlation model that indicates a correlation between a plurality of parameters indicating the properties of each of a plurality of kerosenes, the plurality of parameters including at least density, distillation properties, and aromatic content, and the amount of naphthenobenzene contained in each of the plurality of kerosenes, a function of acquiring numerical values of the plurality of parameters of a specific kerosene, and a function of predicting the amount of naphthenobenzene contained in the specific kerosene using the model data and the numerical values of the plurality of parameters of the specific kerosene.

Yet another aspect of the present disclosure is an information processing device. This device includes a storage unit that stores model data relating to a correlation model that shows a correlation between a plurality of parameters indicating the properties of each of a plurality of kerosenes, the plurality of parameters including at least density, distillation properties, and aromatic content, and the amount of naphthenobenzene contained in each of the plurality of kerosenes, and stores base oil data including numerical values of the plurality of parameters of each of a plurality of base oils used in the production of a target kerosene, and a calculation unit that uses the model data and the base oil data to determine a target blending ratio of the plurality of base oils for producing the target kerosene so that the amount of naphthenobenzene contained in the target kerosene is equal to or less than a predetermined reference value.

This disclosure makes it possible to easily predict the amount of naphthenobenzene contained in kerosene.

FIG. 1 is a diagram illustrating a schematic configuration of an information processing device according to an embodiment. FIG. 2 is a diagram showing an example of a data structure of base oil data. FIG. 2 is a diagram showing an example of a data structure of fuel oil data. 1 is a graph showing an example of a prediction result of an amount of naphthenobenzene. 1 is a graph showing an example of the relationship between the kinetic viscosity and density of kerosene and the combustibility. 1 is a flowchart showing a method for predicting an amount of naphthenobenzene according to an embodiment. 4 is a flowchart showing a method for determining a blend ratio according to an embodiment.

The present disclosure provides an overview. The present disclosure relates to a technology for predicting the amount of naphthenobenzene contained in kerosene from the numerical values of multiple parameters that indicate the properties of the kerosene. Measuring the amount of naphthenobenzene requires the use of mass spectrometry such as field ionization mass spectrometry (FIMS), gas chromatography time-of-flight mass spectrometry (GC-TOFMS), and two-dimensional gas chromatography mass spectrometry (2DGC), which is cumbersome and time-consuming. Therefore, it is not realistic to sequentially measure the amount of naphthenobenzene in kerosene at the kerosene production site. In the present disclosure, the amount of naphthenobenzene is predicted using a correlation model that inputs multiple parameters that can be obtained at the kerosene production site. The multiple parameters used to predict the amount of naphthenobenzene can include density, kinetic viscosity, distillation properties, and aromatic content.

The present disclosure relates to a technique for determining a target blending ratio when blending multiple base oils to produce a target kerosene. According to the present disclosure, even if the naphthenobenzene amount in the base oil is unknown, the naphthenobenzene amount can be predicted from multiple parameters of the base oil using a correlation model. As a result, by using the correlation model, it is possible to determine the blending ratio of multiple base oils so that the naphthenobenzene amount in the blended target kerosene is equal to or less than a predetermined reference value. This makes it possible to easily determine the blending ratio for producing a target kerosene with excellent long-term combustibility.

The technology of the present disclosure will be described below with reference to the drawings based on preferred embodiments. The embodiments are illustrative and do not limit the invention, and all features and combinations thereof described in the embodiments are not necessarily essential to the invention. The same or equivalent components, members, and processes shown in each drawing are given the same reference numerals, and duplicated descriptions are omitted as appropriate. In addition, the scale and shape of each part shown in each drawing are set for convenience to facilitate explanation, and are not to be interpreted as being limiting unless otherwise specified. In addition, when terms such as "first" and "second" are used in this specification or claims, they do not represent any order or importance, but are intended to distinguish one configuration from another, unless otherwise specified.

The subject of the device or method of the present disclosure includes a computer. The computer executes a computer program to realize the functions of the subject of the device or method of the present disclosure. The computer includes, as its main hardware configuration, a processor that operates according to the computer program. The type of processor is not important as long as it can realize the functions by executing the computer program. The processor is composed of one or more electronic circuits including a semiconductor integrated circuit (IC, LSI, etc.). The computer program is recorded on a non-transitory recording medium such as a computer-readable ROM, optical disk, or hard disk drive. The computer program may be stored in advance on the recording medium, or may be supplied to the recording medium via a wide area communication network including the Internet.

FIG. 1 is a diagram showing a schematic functional configuration of an information processing device 10 according to an embodiment. The information processing device 10 includes an acquisition unit 12, a storage unit 14, a calculation unit 16, and an output unit 18. The information processing device 10 is connected to a user terminal 50 and a local device 52 via a network 40.

Each block shown in the block diagram of this disclosure can be realized in terms of hardware by elements or mechanical devices, including a processor such as a computer's CPU (Central Processing Unit) and memories such as ROM (Read Only Memory) and RAM (Random Access Memory), and in terms of software by computer programs, etc. Here, functional blocks realized by the cooperation of hardware and software are depicted. Those skilled in the art will understand that these functional blocks can be realized in various ways by combining hardware and software.

A computer program that implements the functions of at least some of the multiple functional blocks shown in FIG. 1 may be installed in the storage of one or more computers. The CPU of one or more computers may load the computer program installed in the computer into main memory and execute it to perform the functions of the multiple functional blocks shown in FIG. 1.

Furthermore, the functions of the multiple functional blocks shown in FIG. 1 may be executed by one computer, or may be executed in a distributed manner by multiple computers. When the functions of the multiple functional blocks shown in FIG. 1 are executed in a distributed manner by multiple computers, the multiple computers may transmit and receive data via a communication network including a LAN (Local Area Network), a WAN (Wide Area Network), and the Internet.

The information processing device 10 is, for example, a server. The information processing device 10 may be a general-purpose computer such as a workstation or a personal computer. The information processing device 10 may be a mobile terminal such as a smartphone or a tablet computer. The information processing device 10 may be a cloud-type server configured to provide computer resources to a user terminal 50 via a network 40.

The user terminal 50 is, for example, a general-purpose computer such as a personal computer. The user terminal 50 may also be a mobile terminal such as a smartphone or a tablet computer. The user terminal 50 is used, for example, in factories, laboratories, etc. that handle base oils and fuel oils.

The local device 52 is, for example, a server. The local device 52 may be a general-purpose computer such as a workstation or a personal computer. The local device 52 is installed, for example, in a factory or laboratory that handles base oil or fuel oil. The local device 52 functions as a data server that centrally manages data used in the factory or laboratory. The local device 52 may be a cloud-type server configured to provide computer resources to the user terminal 50 via the network 40. In this case, the local device 52 may be installed at a base away from the factory or laboratory where the user terminal 50 is installed.

Network 40 is composed of a communication network including, for example, a LAN (Local Area Network), a WAN (Wide Area Network), the Internet, and various mobile communication systems constructed by wireless base stations. Examples of mobile communication systems include mobile communication systems such as 3G, 4G, and 5G, LTE (Long Term Evolution), and wireless networks (for example, Wi-Fi (registered trademark)) that can be connected to the Internet via a specified access point.

The acquisition unit 12 acquires various data from the user terminal 50 or the local device 52 via the network 40. The acquisition unit 12 may acquire data input by the user through any input device, such as a keyboard or a mouse, connected to the information processing device 10.

The memory unit 14 stores data necessary for the processing executed by the calculation unit 16. The memory unit 14 records and manages the data acquired by the acquisition unit 12. The memory unit 14 stores base oil data 20, kerosene data 22, and model data 24. The memory unit 14 is, for example, a database that accumulates data related to base oil and kerosene, and stores past historical data related to base oil and kerosene produced in factories and laboratories.

Figure 2 is a diagram showing an example of the data structure of base oil data 20. Base oil data 20 is data related to base oil, which is a raw material used in the production of kerosene. Base oil data 20 includes an ID and name for identifying the base oil, and multiple parameters that indicate the properties of the base oil. Base oil data 20 includes multiple parameters that can be obtained at the base oil production site.

The base oil data 20 includes density (15°C, g/ ^cm3 ) and kinematic viscosity (30°C, ^mm2 /s). The density at 15°C can be measured in accordance with JIS K2249 "Crude oil and petroleum products-Determination of density". The kinematic viscosity at 30°C can be measured in accordance with JIS K2283 "Crude oil and petroleum products-Test method for kinematic viscosity and calculation method for viscosity index".

The base oil data 20 includes sulfur content (ppm) and calorific value (kJ/kg). The sulfur content can be measured in accordance with JIS K2541 "Crude oil and petroleum products - Testing method for sulfur content". The calorific value can include at least one of gross calorific value and net calorific value. The calorific value can be obtained in accordance with JIS K2279 "Crude oil and petroleum products - Testing method for calorific value and method for estimating calorific value by calculation".

The base oil data 20 may include a distillation temperature (°C) as a distillation property, for example, an initial boiling point temperature (IPB), a vol% distillation temperature (Ta) and an end temperature (EP). The base oil data 20 may include at least one of the following a vol% distillation temperature: 5 vol% distillation temperature (T5), 10 vol% distillation temperature (T10), 20 vol% distillation temperature (T20), 30 vol% distillation temperature (T30), 40 vol% distillation temperature (T40), 50 vol% distillation temperature (T50), 60 vol% distillation temperature (T60), 70 vol% distillation temperature (T70), 80 vol% distillation temperature (T80), 90 vol% distillation temperature (T90), 95 vol% distillation temperature (T95) and 97 vol% distillation temperature (T97). The distillation temperature can be measured in accordance with JIS K2254 "Petroleum products - Determination of distillation properties", for example, using a simplified distillation method.

The base oil data 20 may include a distillation percentage (volume %) of b°C or higher as a distillation property. The percentage of distillation of b°C or higher (%) refers to the volume fraction of distillation at a temperature of b°C or higher. The base oil data 20 may include at least one of the following as the percentage of distillation of b°C or higher (Zb): a percentage of distillation of 210°C or higher (Z210), a percentage of distillation of 235°C or higher (Z235), and a percentage of distillation of 240°C or higher (Z240). The percentage of distillation of b°C or higher (Zb) may be obtained, for example, by obtaining a distillation curve from the measured value of the distillation temperature and calculating the volume fraction of the distillation curve that is b°C or higher. The percentage of distillation of b°C or higher (Zb) may be obtained by subtracting the percentage of distillation of b°C or lower (Eb) from 100% (i.e., Zb = 100 - Eb). The base oil data 20 may include the fraction distilled at b°C or less (Eb) as a distillation property, and may include, for example, at least one of the fraction distilled at 210°C or less (E210), the fraction distilled at 235°C or less (E235), and the fraction distilled at 240°C or less (E240).

The base oil data 20 can include component ratios (volume percent) of saturates, olefins, total aromatics, one-ring aromatics, two-ring aromatics, and three or more ring (3+) aromatics. These component ratios can be measured, for example, in accordance with the Japan Petroleum Institute method JPI-5S-49-97 "Petroleum products - Hydrocarbon type test method - High performance liquid chromatography."

The type of base oil recorded in the base oil data 20 is not particularly limited. The base oil may be a kerosene fraction obtained from a crude oil atmospheric distillation unit or a vacuum distillation unit, or may be a kerosene fraction obtained by at least one of hydrodesulfurization, catalytic cracking, hydrocracking, and thermal cracking. The base oil may be, for example, at least one of straight-run kerosene obtained from a crude oil atmospheric distillation unit, a kerosene fraction obtained by hydrodesulfurizing straight-run kerosene, a kerosene fraction obtained by catalytic cracking of heavy diesel or residual oil (straight-run heavy oil) obtained from an atmospheric distillation unit, a kerosene fraction obtained by thermal cracking of straight-run heavy oil, a kerosene fraction obtained by hydrocracking straight-run heavy oil, a kerosene fraction obtained by hydrocracking vacuum diesel obtained from a vacuum distillation unit, a kerosene fraction obtained by hydrocracking heavy diesel mixed with catalytic cracked diesel, a kerosene fraction obtained by desulfurizing catalytic cracked diesel, and a desulfurized kerosene fraction derived from oil sand.

The base oil may be a residue (raffinate) after extracting a specific component (e.g., a light fraction) from a kerosene fraction. The base oil may be a nor-para raffinate, which is a residual oil after extracting a normal paraffin fraction from a kerosene fraction. The base oil may be a residue after extracting a light fraction having a distillation temperature of 160°C or more and 190°C or less from a kerosene fraction. The base oil may be a C9 aromatic hydrocarbon mainly composed of aromatic hydrocarbons having 9 carbon atoms. The C9 aromatic hydrocarbon may be, for example, a residue after extracting C6 to C8 aromatic hydrocarbons from C6 to C9 aromatic hydrocarbons having 6 to 9 carbon atoms. The base oil may be a trimer or higher of isobutylene or isooctene heavy obtained by hydrogenating it.

The base oil may be a non-petroleum-derived base oil. The base oil may be an FT synthetic oil obtained by Fischer-Tropsch synthesis (FT synthesis) from a mixed gas of carbon monoxide and hydrogen, or a kerosene fraction obtained by fractionating an FT synthetic oil, a kerosene fraction obtained by hydrogenating an FT synthetic oil, or a kerosene fraction obtained by isomerizing an FT synthetic oil. The base oil may be a bio-derived oil obtained from plants or bacteria, or a kerosene fraction obtained by fractionating an oil derived from a bio-derived material, a kerosene fraction obtained by hydrogenating an oil derived from a bio-derived material, or a kerosene fraction obtained by isomerizing an oil derived from a bio-derived material. The base oil may be a waste oil-derived recycled oil made from waste oil such as waste plastics, waste tires, or waste edible oil, or a kerosene fraction obtained by hydrogenating a waste oil-derived recycled oil, or a kerosene fraction obtained by isomerizing a waste oil-derived recycled oil.

The base oil data 20 may further include information for identifying the base oil, such as the type name of the base oil, the place of production, the date and time of production, and the serial number. The base oil data 20 may further include information regarding the cost of the base oil, and may include the price per unit volume for producing the base oil. The base oil data 20 may further include information regarding the inventory amount of the base oil, and may include the volume of base oil stored at a production plant or the like.

Figure 3 is a diagram showing an example of the data structure of kerosene data 22. Kerosene data 22 is data about kerosene produced by blending multiple base oils. Kerosene data 22 includes an ID and name for identifying the kerosene, and multiple parameters that indicate the properties of the kerosene.

The kerosene data 22 has parameters in common with the base oil data 20. Like the base oil data 20, the kerosene data 22 can have density, kinetic viscosity, sulfur content, calorific value, distillation temperature, distillation ratio, and component ratio. The kerosene data 22 has a parameter different from the base oil data 20, namely, the amount of naphthenobenzene. The kerosene data 22 can have the ratio (volume %) of 1-ring naphthenobenzene and 2-ring naphthenobenzene as the amount of naphthenobenzene. The amount of naphthenobenzene (volume %) of kerosene may be the sum of the amount of 1-ring naphthenobenzene (volume %) and the amount of 2-ring naphthenobenzene (volume %). The amount of naphthenobenzene can be measured using mass spectrometry such as FIMS, GC-TOFMS, and 2DGC.

The kerosene data 22 may further include component ratios of any component that can be analyzed using mass spectrometry. The kerosene data 22 may further include component ratios of normal paraffins, isoparaffins, 1-ring cycloparaffins, 2-ring cycloparaffins, 3-ring cycloparaffins, 4-ring cycloparaffins, alkylbenzenes, naphthalenes, acenaphthenes, biphenyls, fluorenes, phenanthrenes, etc.

Kerosene data 22 may further include information for identifying the kerosene, such as the type of kerosene, the place of production, the date and time of production, and the serial number. Kerosene data 22 may further include information regarding the cost of the kerosene, and may include the price per unit volume for producing the kerosene.

ここで、関数ｆ_ｉは、パラメータｘ_ｉを変数とする任意の関数であり、例えば、パラメータｘ_ｉの多項式である。係数ｐは、切片である。 Returning to Fig. 1, the model data 24 includes various parameters necessary for constructing a correlation model for predicting the amount of naphthenobenzene in kerosene. The correlation model represents the correlation between the numerical values x _i of multiple parameters indicating the properties of kerosene and the amount of naphthenobenzene y in kerosene. The correlation model can be represented, for example, by the following formula (1).

Here, the function f _i is an arbitrary function with the parameter x _i as a variable, for example, a polynomial of the parameter x _i , and the coefficient p is an intercept.

ここで、係数ｋ_ｉは、パラメータｘ_ｉに対する偏回帰係数であり、係数ｋ_ｐは、切片である。 The correlation model expressed by the formula (1) may be a nonlinear model or a linear model. When the correlation model is linear, the correlation model can be expressed by, for example, the following formula (23).

Here, the coefficient k _i is the partial regression coefficient for the parameter x _i , and the coefficient k _p is the intercept.

As the multiple parameters x _i included in the correlation model, multiple parameters included in the base oil data 20 in Fig. 2 and the kerosene data 22 in Fig. 3 can be used. The multiple parameters x _i include at least density, distillation properties, and aromatic content.

The multiple parameters x _i preferably include at least one distillation temperature Ta as a distillation property, and preferably include at least one of 50% distillation temperature (T50), 70% distillation temperature (T70), and 90% distillation temperature (T90). The multiple parameters x _i more preferably include multiple distillation temperatures Ta with different volume fractions a, and preferably include, for example, T50 and T70. The multiple parameters x _i may include T50, T70, and T90.

The multiple parameters x _i preferably include at least one distillation ratio (Zb) of b ° C or more as a distillation property, and preferably include at least one of a distillation ratio of 210 ° C or more (Z210), a distillation ratio of 235 ° C or more (Z235), and a distillation ratio of 240 ° C or more (Z240). The multiple parameters x _i may include multiple distillation ratios Zb of b ° C or more with different temperatures b, for example, Z235 and Z240.

The multiple parameters x _i preferably include a total aromatic content and a polycyclic aromatic content as the aromatic content. The polycyclic aromatic content can be calculated by adding up the two-ring aromatic content and the three-ring + aromatic content. The multiple parameters x _i may include a total aromatic content and a one-ring aromatic content as the aromatic content. The multiple parameters x _i may include a one-ring aromatic content and a polycyclic aromatic content as the aromatic content.

In one embodiment, the number of multiple parameters x _i used as inputs to the correlation model is set to 6 or more, and at least density, T50, T70, total aromatic content, and polycyclic aromatic content are used, so that the ^Q2 value, which is an index of the prediction accuracy of the correlation model, can be set to 0.7 or more.

Specific values of various parameters (e.g., coefficients k _i , k _p ) constituting the correlation model can be determined by machine learning. For example, the kerosene data 22 in FIG. 3 can be used as teacher data in machine learning. Specifically, the correlation model can be constructed by supervised learning in which the multiple parameters x _i of each of the multiple kerosenes included in the kerosene data 22 are input and the naphthenobenzene amount y is output. At least one of linear regression, support vector machine, random forest, gradient boosting, and neural network can be used as the machine learning algorithm. The correlation model does not have to be generated by machine learning, and may be generated by a method different from machine learning. The correlation model may be a lookup table that defines the relationship between the input value of the multiple parameters x _i and the output value of the naphthenobenzene amount y.

The calculation unit 16 includes a model generation unit 30, a prediction unit 32, and a blending determination unit 34.

The model generation unit 30 executes machine learning to generate the model data 24. For example, the model generation unit 30 uses the kerosene data 22 stored in the storage unit 14 as teacher data to generate a correlation model for predicting the naphthenobenzene amount in kerosene and determines the values of various parameters constituting the correlation model. The model data 24 generated by the model generation unit 30 is stored in the storage unit 14. The model generation unit 30 may generate the model data 24 by a method other than machine learning. The model data 24 may be a lookup table that defines the relationship between the input values of multiple parameters x _i and the output value of the naphthenobenzene amount y.

The prediction unit 32 predicts the amount of naphthenobenzene in a specific kerosene blended at a specific blending ratio of multiple base oils, using the model data 24 stored in the memory unit 14. The prediction unit 32 inputs, for example, the numerical values of multiple parameters x _i indicating the properties of the specific kerosene into a correlation model, and sets the amount of naphthenobenzene y output from the correlation model as a predicted value.

The prediction unit 32 may predict the naphthenobenzene amount y of the specific kerosene using the numerical values of multiple parameters x _i indicating the properties of the specific kerosene input to the user terminal 50 and acquired by the acquisition unit 12. A user using the user terminal 50 may input the analysis results of a newly blended or produced specific kerosene to the user terminal 50. A user using the user terminal 50 may calculate the numerical values of multiple parameters x _i of the specific kerosene by taking a weighted average of multiple parameters x _i indicating the properties of each of multiple base oils used in the production of the specific kerosene with the blending ratio, and input the calculated values to the user terminal 50.

When calculating the distillation properties of a specific kerosene by weighted averaging from the distillation properties of each of multiple base oils, it is important to note that simple additivity does not hold for distillation properties by simple distillation. It is known that simple additivity holds for distillation properties by true boiling point distillation rather than simple distillation. Therefore, when the distillation properties of the base oil data 20 are by simple distillation, the distillation temperature Ta by the simple distillation is converted to the distillation temperature Tc by true boiling point distillation, and the weighted average is taken with the distillation temperature Tc by the true boiling point distillation. After that, the weighted average distillation temperature Tc is converted back to the distillation temperature Ta by the simple distillation, so that the distillation properties of a specific kerosene by the simple distillation can be calculated. As a method for converting the distillation temperatures Ta and Tc between the simple distillation method and the true boiling point distillation method, for example, a known method described in API Technical Databook Fifth Edition can be used.

FIG. 4 is a graph showing an example of the prediction result of the naphthenobenzene amount. FIG. 4 shows the relationship between the actual value and the predicted value of the naphthenobenzene amount of various kerosene. The actual value of the naphthenobenzene amount on the horizontal axis is a value measured by mass spectrometry (for example, GC-TOFMS). The predicted value of the naphthenobenzene amount on the vertical axis is a value predicted by the prediction unit 32 using the correlation model. The solid line 60 in FIG. 4 is a straight line with a slope of 1 where the actual value and the predicted value match. The example of FIG. 4 shows a case where linear regression (for example, multiple regression analysis) is used as a machine learning algorithm for generating the correlation model. The example of FIG. 4 shows a case where density, T50, T70, T90, total aromatic content, polycyclic aromatic content, and Z210 are used as multiple parameters x _i .

As shown in Fig. 4, the plot of the naphthenobenzene amount is distributed near the solid line 60, and it is understood that the actual measured value can be accurately predicted. In Fig. 4, the error on the negative side indicated by the dashed line 62 is 1.079%, and the error on the positive side indicated by the dashed line 64 is 1.396%. In Fig. 4, the ^Q2 value, which is an index of the prediction accuracy of the correlation model, is 0.755, and it is understood that the naphthenobenzene amount can be predicted with high accuracy.

In addition, since the amount of naphthenobenzene contained in kerosene is required to be equal to or less than a predetermined reference value (e.g., 8.9% by volume), it is undesirable for the predicted value of the naphthenobenzene amount to be lower than the actual measured value. Therefore, the prediction unit 32 may set the predicted value of the naphthenobenzene amount to a value obtained by adding the negative error value in the correlation model (e.g., 1.079%) as a safety factor. In this case, the prediction unit 32 can calculate the predicted value so that the predicted value of the naphthenobenzene amount always exceeds the actual measured value.

The blending determination unit 34 uses the base oil data 20 and model data 24 stored in the memory unit 14 to determine a target blending ratio of multiple base oils for producing a target kerosene in which the amount of naphthenobenzene is equal to or less than a predetermined reference value. The blending determination unit 34 determines the target blending ratio so that the predicted value of the amount of naphthenobenzene in the target kerosene blended at the target blending ratio is equal to or less than the predetermined reference value. The blending determination unit 34 can determine the target blending ratio of multiple base oils by using an optimization calculation method such as the simulated annealing method. The blending determination unit 34 determines, for example, an ID for identifying the multiple base oils used in the blending and the blending ratio (volume %) of each of the multiple base oils.

The blending determination unit 34 may determine the target blending ratio of the multiple base oils, for example, using a reference value of the naphthenobenzene amount input to the user terminal 50 and acquired by the acquisition unit 12. The blending determination unit 34 may determine the target blending ratio of the multiple base oils, using a reference value of the naphthenobenzene amount stored in advance in the memory unit 14. The reference value of the naphthenobenzene amount is, for example, 8.9% by volume. By setting the naphthenobenzene amount contained in the target kerosene to 8.9% by volume or less, deterioration of the combustibility of the target kerosene produced according to the calculated target blending ratio can be suppressed.

The blending determination unit 34 may determine the target blending ratio of the multiple base oils using constraints other than the amount of naphthenobenzene. The blending determination unit 34 may determine the target blending ratio so that the distillation properties of the target kerosene are in a desired numerical range. The blending determination unit 34 may determine the target blending ratio so that the difference between the 10% by volume distillation temperature (T10) and the 50% by volume distillation temperature (T50) of the target kerosene is greater than 10°C (i.e., T50-T10>10°C). By making the target kerosene satisfy the condition T50-T10>10°C, the deterioration of the fuel properties of the target kerosene produced according to the calculated target blending ratio can be suppressed. If the distillation temperature range is too narrow, the light components may vaporize all at once in a combustion test, resulting in an over-fuel state and a deterioration in combustibility.

The blending determination unit 34 may determine the target blending ratio so that the difference between the a volume % distillation temperature and the (a+25) volume % distillation temperature of the target kerosene is 7°C or more. For example, the target blending ratio may be determined so that a = 5 and the difference between the 5 volume % distillation temperature (T5) and the 30 volume % distillation temperature (T30) is 7°C or more (i.e., T30-T5≧7°C). For example, the target blending ratio may be determined so that a = 70 and the difference between the 70 volume % distillation temperature (T70) and the 95 volume % distillation temperature (T95) is 7°C or more (i.e., T95-T70≧7°C). In addition, the target blending ratio may be determined so that multiple conditions (e.g., a = 5 and a = 70) are satisfied simultaneously. This prevents the distillation temperature range of the target kerosene produced according to the calculated target blending ratio from narrowing, and prevents the fuel properties of the target kerosene from deteriorating.

The blending determination unit 34 may determine the target blending ratio so that the difference between the a volume % distillation temperature and the (a+20) volume % distillation temperature of the target kerosene is 6°C or more. For example, the target blending ratio may be determined so that a = 30 and the difference between the 30 volume % distillation temperature (T30) and the 50 volume % distillation temperature (T50) is 6°C or more (i.e., T50-T30 ≧ 6°C). For example, the target blending ratio may be determined so that a = 50 and the difference between the 50 volume % distillation temperature (T50) and the 70 volume % distillation temperature (T70) is 6°C or more (i.e., T70-T50 ≧ 6°C). In addition, the target blending ratio may be determined so that multiple conditions (e.g., a = 30 and a = 50) are satisfied simultaneously. This prevents the distillation temperature range of the target kerosene produced according to the calculated target blending ratio from narrowing, and prevents the fuel properties of the target kerosene from deteriorating.

The blending determination unit 34 may determine the target blending ratio so that the kinetic viscosity v (30° C., mm ² /s) and density ρ (15° C., g/cm ³ ) of the target kerosene satisfy predetermined conditions. The blending determination unit 34 may determine the target blending ratio so that the conditions v≦1.522, ρ≦0.8061, and ρ≦−0.0429v+0.08653 are satisfied. This makes it possible to suppress deterioration of the fuel properties of the target kerosene produced according to the calculated target blending ratio. If the kinetic viscosity or density is too high, there is a possibility that poor volatilization or poor combustion will occur in a combustion test, resulting in poor combustibility.

Figure 5 is a graph showing an example of the relationship between the kinetic viscosity ν and density ρ of kerosene and the flammability. In Figure 5, a plot 72 (○) where the flammability test passes and a plot 74 (×) where the flammability test fails are arranged on a graph with the kinetic viscosity ν on the horizontal axis and the density ρ on the vertical axis for various kerosene. The shaded area 70 in Figure 5 is a range that satisfies the conditions ν≦1.522, ρ≦0.8061, and ρ≦-0.0429ν+0.08653. As shown in Figure 5, by determining the target blending ratio so that the kinetic viscosity ν and density ρ are included in the shaded area 70, it is possible to prevent the target kerosene produced according to the calculated target blending ratio from failing the fuel property test. For example, when a 1000-hour long-term combustion evaluation test is conducted on a heating device such as a kerosene stove, the flammability test passes if no defects occur, such as a large flame, backfire, visible smoke, deterioration of the wick in a wick stove, deterioration of the nozzle in a fan heater, oil leaks or damage, a decrease in flow rate due to carbon accumulation on the wick or nozzle causing a small flame, or a deterioration in exhaust gases such as an increase in carbon monoxide (CO).

The blending determination unit 34 may determine the target blending ratio so that the component ratio of a specific component contained in the target kerosene blended at the target blending ratio is equal to or less than a predetermined reference value. As the specific component, a component whose reference value is set by law or standard specification may be used, and for example, a reference value of 0.0080 mass% or less of sulfur content may be used.

The blending determination unit 34 may determine the target blending ratio so that the calorific value (total calorific value or net calorific value) of the target kerosene blended at the target blending ratio satisfies a specified condition. The calorific value of the target kerosene blended at the target blending ratio can be calculated using the calorific values of multiple base oils included in the base oil data 20. For example, the calorific value of the target kerosene can be calculated by weighting the calorific values of each of the multiple base oils by the target blending ratio, and it can be confirmed whether the calculated calorific value of the target kerosene satisfies the specified condition. Here, the specified condition may be a condition in which the calorific value of the target kerosene is within a specified numerical range, or a condition in which the calorific value of the target kerosene is equal to or greater than a specified lower limit.

The blending determination unit 34 may determine the target blending ratio so that the cost of the target kerosene blended at the target blending ratio satisfies a specified condition. The cost of the target kerosene blended at the target blending ratio can be calculated using the costs of multiple base oils included in the base oil data 20. For example, the cost of the target kerosene can be calculated by weighting the costs of the multiple base oils by the target blending ratio, and it can be confirmed whether the calculated cost of the target kerosene satisfies the specified condition. Here, the specified condition may be a condition in which the cost of the target kerosene is within a specified numerical range, or a condition in which the cost of the target kerosene is less than a specified upper limit.

The blending determination unit 34 may determine the target blending ratio so that the target kerosene blended at the target blending ratio is produced within the range of the inventory amounts of the multiple base oils. In this case, the blending determination unit 34 further uses, for example, the target production amount of the target kerosene input to the user terminal 50 and acquired by the acquisition unit 12. The blending determination unit 34 calculates the usage amount of each of the multiple base oils using the target production amount of the target kerosene and the target blending ratios of the multiple base oils, and determines the target blending ratio so that the usage amount of each of the multiple base oils is equal to or less than the inventory amount. The blending determination unit 34 can use the inventory amount included in the base oil data 20.

The output unit 18 outputs the processing results of the calculation unit 16. The output unit 18 may output the processing results to a user terminal 50 via the network 40, or may output the processing results to an output device such as a display connected to the information processing device 10. The output unit 18 outputs the naphthenobenzene amount of a specific kerosene predicted by the prediction unit 32, and the target blending ratio of multiple base oils determined by the blending determination unit 34.

FIG. 6 is a flowchart showing a method for predicting the amount of naphthenobenzene according to an embodiment. The prediction unit 32 acquires model data relating to a correlation model showing the correlation between a plurality of parameters indicating the properties of each of a plurality of kerosenes and the amount of naphthenobenzene contained in each of the plurality of kerosenes (step S10). The plurality of parameters include at least density, distillation properties, and aromatic content, and preferably includes, for example, density, T50, T70, total aromatic content, and polycyclic aromatic content. The acquisition unit 12 acquires numerical values of a plurality of parameters of a specific kerosene (step S12). The numerical values of the plurality of parameters of a specific kerosene may be measured by analyzing the specific kerosene, or may be calculated by a method such as weighted averaging from a plurality of parameters measured for a plurality of base oils used in the production of the specific kerosene. The prediction unit 32 predicts the amount of naphthenobenzene contained in the specific kerosene using the model data and the numerical values of the plurality of parameters of the specific kerosene (step S14). By inputting the numerical values of the plurality of parameters into the correlation model, a predicted value of the amount of naphthenobenzene is output from the correlation model. The order of steps S10 and S12 is not particularly limited, and the order of steps S10 and S12 may be reversed.

Figure 7 is a flowchart showing a method for determining a blending ratio according to an embodiment. The blending determination unit 34 acquires model data relating to a correlation model showing a correlation between a plurality of parameters indicating the properties of each of the plurality of kerosenes and the amount of naphthenobenzene contained in each of the plurality of kerosenes (step S20). The plurality of parameters include at least density, distillation properties, and aromatic content, and preferably includes, for example, density, T50, T70, total aromatic content, and polycyclic aromatic content. The blending determination unit 34 acquires base oil data including numerical values of a plurality of parameters of each of the plurality of base oils used in the production of the target kerosene (step S22). Using the model data and the base oil data, the blending determination unit 34 determines the blending ratio of the plurality of base oils for producing a target kerosene in which the amount of naphthenobenzene contained in the target kerosene is equal to or less than a predetermined reference value (step S24). The order of steps S20 and S22 is not particularly limited, and the order of S20 and S22 may be reversed.

According to this embodiment, even if the naphthenobenzene amount of a specific kerosene cannot be measured at the kerosene production site, the naphthenobenzene amount can be accurately estimated from parameters that can be obtained at the kerosene production site. According to this embodiment, the blending ratio of multiple base oils for producing a target kerosene can be determined so that the naphthenobenzene amount of the target kerosene is equal to or less than a predetermined reference value. As a result, even when a base oil estimated to have a high naphthenobenzene content is used as the base oil, the naphthenobenzene amount can be prevented from becoming excessive, and deterioration of the combustibility of the target kerosene produced according to the target blending ratio can be suppressed.

According to this embodiment, base oils estimated to have a high naphthenobenzene content can be actively used, making it easier to increase production of kerosene with excellent combustibility. The first base oil having a relatively high naphthenobenzene content can be selected from a first group including straight-run kerosene, a kerosene fraction obtained by hydrodesulfurizing straight-run kerosene, a kerosene fraction obtained by hydrocracking straight-run heavy oil, a kerosene fraction obtained by hydrocracking catalytic cracking diesel at a low cracking rate, a kerosene fraction obtained by fractionating and desulfurizing catalytic cracking diesel, a desulfurized kerosene fraction derived from oil sands, and a residual oil obtained by distilling and separating the 160°C to 190°C fraction from the kerosene fraction. In addition, the second base oil having a relatively low naphthenobenzene content is selected from the second group including, for example, norpara raffinate, kerosene fraction obtained by hydrocracking catalytic cracking gas oil at a high cracking rate, C9 aromatic hydrocarbons, isooctene heavy, FT synthetic oil, kerosene fraction obtained by fractionating FT synthetic oil, kerosene fraction obtained by hydrogenating FT synthetic oil, kerosene fraction obtained by isomerizing FT synthetic oil, bio-derived oil, kerosene fraction obtained by fractionating bio-derived oil, kerosene fraction obtained by hydrogenating bio-derived oil, kerosene fraction obtained by isomerizing bio-derived oil, waste oil-derived recycled oil, kerosene fraction obtained by fractionating waste oil-derived recycled oil, kerosene fraction obtained by hydrogenating waste oil-derived recycled oil, and kerosene fraction obtained by isomerizing waste oil-derived recycled oil. By mixing such a first base oil and a second base oil to produce kerosene, it becomes easy to increase the production of kerosene in which the amount of naphthenobenzene is equal to or less than a predetermined standard value.

According to this embodiment, by taking into consideration the distillation properties of the base oil, base oils that are estimated to lead to deterioration of combustibility due to their narrow distillation properties (temperature range) are actively used, making it easier to increase production of kerosene with excellent combustibility. The first base oil having a relatively narrow distillation property (temperature range) is selected from a first group including kerosene fraction obtained by hydrocracking catalytic cracking diesel at a high cracking rate, C9 aromatic hydrocarbons, isooctene heavy, bio-based raw material derived oil, kerosene fraction obtained by fractionating bio-based raw material derived oil, kerosene fraction obtained by hydrogenating bio-based raw material derived oil, kerosene fraction obtained by isomerizing bio-based raw material derived oil, waste oil derived recycled oil, kerosene fraction obtained by fractionating waste oil derived recycled oil, kerosene fraction obtained by hydrogenating waste oil derived recycled oil, and kerosene fraction obtained by isomerizing waste oil derived recycled oil. In addition, the second base oil having a relatively narrow distillation property (temperature range) is selected from a second group including a kerosene fraction obtained by hydrodesulfurizing straight-run kerosene, a kerosene fraction obtained by hydrocracking straight-run heavy oil, a kerosene fraction obtained by hydrocracking catalytic cracking light oil at a low cracking rate, a kerosene fraction obtained by fractionating and desulfurizing catalytic cracking light oil, a desulfurized kerosene fraction derived from oil sands, a residual oil obtained after extracting a fraction having a temperature of 160°C or higher and 190°C or lower from a kerosene fraction, a nor-para raffinate, an FT synthetic oil, a kerosene fraction obtained by fractionating an FT synthetic oil, a kerosene fraction obtained by hydrogenating an FT synthetic oil, and a kerosene fraction obtained by isomerizing an FT synthetic oil. By mixing such a first base oil and a second base oil to produce kerosene, it becomes easier to increase the production of kerosene having a not too narrow distillation property (temperature range).

The present disclosure has been described above based on an embodiment. This embodiment is merely an example, and it will be understood by those skilled in the art that various modifications are possible in the combination of each component or each processing process, and that such modifications are also within the scope of the present disclosure.

In the above embodiment, the calculation unit 16 includes the model generation unit 30, the prediction unit 32, and the combination determination unit 34. In another embodiment, the calculation unit 16 does not need to include at least any of the model generation unit 30, the prediction unit 32, and the combination determination unit 34. The model data 24 used by the prediction unit 32 or the combination determination unit 34 does not need to be generated by the model generation unit 30, and may be generated in advance in a device different from the information processing device 10.

The embodiment may be a program for causing a computer to realize the functions for implementing the above-mentioned method, or a recording medium for storing the program. Such a recording medium for storing the program may be a non-transitory, tangible, computer-readable storage medium, and may be a non-volatile memory, a magnetic recording medium such as a magnetic tape or a magnetic disk, or an optical recording medium such as an optical disk.

Several aspects of this disclosure are described below.

The first aspect of the present disclosure is an information processing device including: a storage unit that stores model data relating to a correlation model that indicates a correlation between a plurality of parameters indicating the properties of each of a plurality of kerosenes, the plurality of parameters including at least density, distillation properties, and aromatic content, and the amount of naphthenobenzene contained in each of the plurality of kerosenes; an acquisition unit that acquires the numerical values of the plurality of parameters of a specific kerosene; and a calculation unit that predicts the amount of naphthenobenzene contained in the specific kerosene using the model data and the numerical values of the plurality of parameters of the specific kerosene. According to the first aspect, the amount of naphthenobenzene, which requires a lot of time and effort to analyze, can be predicted using parameters that can be acquired at the kerosene production site. In particular, by using at least density, distillation properties, and aromatic content as the plurality of parameters, the amount of naphthenobenzene contained in the specific kerosene can be predicted with high accuracy.

A second aspect of the present disclosure is the information processing device according to the first aspect, in which the distillation properties include at least one of a 50% distillation temperature, a 70% distillation temperature, and a 90% distillation temperature. According to the second aspect, by using at least one of a 50% distillation temperature, a 70% distillation temperature, and a 90% distillation temperature as a parameter used for prediction, it is possible to improve the prediction accuracy of the amount of naphthenobenzene contained in a specific kerosene.

A third aspect of the present disclosure is the information processing device according to the first or second aspect, in which the distillation properties include at least one of a distillation percentage of 210°C or higher, a distillation percentage of 235°C or higher, and a distillation percentage of 240°C or higher. According to the third aspect, by using at least one of a distillation percentage of 210°C or higher, a distillation percentage of 235°C or higher, and a distillation percentage of 240°C or higher as a parameter used for prediction, it is possible to improve the prediction accuracy of the naphthenobenzene amount contained in a specific kerosene.

A fourth aspect of the present disclosure is the information processing device according to any one of the first to third aspects, in which the aromatic content includes a total aromatic content and a polycyclic aromatic content. According to the fourth aspect, by using both the total aromatic content and the polycyclic aromatic content as parameters used in the prediction, it is possible to improve the prediction accuracy of the amount of naphthenobenzene contained in a specific kerosene.

A fifth aspect of the present disclosure is the information processing device according to any one of the first to fourth aspects, in which the plurality of parameters further includes kinetic viscosity. According to the fifth aspect, by further using kinetic viscosity as a parameter used in the prediction, it is possible to improve the prediction accuracy of the amount of naphthenobenzene contained in a specific kerosene.

A sixth aspect of the present disclosure is the information processing device according to any one of the first to fifth aspects, in which the model data is generated by machine learning using teacher data having numerical values of the plurality of parameters of the plurality of kerosenes as input and the amount of naphthenobenzene contained in the plurality of kerosenes as output. According to the sixth aspect, by generating a correlation model by machine learning, it is possible to improve the prediction accuracy of the amount of naphthenobenzene contained in a specific kerosene.

A seventh aspect of the present disclosure is an information processing device according to any one of the first to sixth aspects, in which the storage unit further stores base oil data having numerical values of the multiple parameters of multiple base oils used in the production of the target kerosene, and the calculation unit uses the model data and the base oil data to determine a target blending ratio of the multiple base oils for producing the target kerosene such that the amount of naphthenobenzene contained in the target kerosene is equal to or less than a predetermined reference value. According to the seventh aspect, it is possible to determine a target blending ratio for producing a target kerosene in which the amount of naphthenobenzene is equal to or less than a predetermined reference value. In particular, since there is no need to analyze the amount of naphthenobenzene contained in each of the multiple base oils used in the production of the target kerosene, the effort of analysis can be significantly reduced.

An eighth aspect of the present disclosure is the information processing device according to the seventh aspect, in which the calculation unit determines the target blending ratio so that the amount of naphthenobenzene contained in the target kerosene is 8.9% by volume or less. According to the eighth aspect, by using a constraint condition that sets the predicted value of the amount of naphthenobenzene contained in the target kerosene to 8.9% by volume or less, it is possible to determine the target blending ratio for producing a target kerosene with excellent long-term combustibility.

A ninth aspect of the present disclosure is the information processing device according to the seventh or eighth aspect, in which the calculation unit determines the target blending ratio so that the difference between the 10% distillation temperature and the 50% distillation temperature of the target kerosene is greater than 10°C. According to the ninth aspect, it is possible to prevent the distillation properties (distillation range) of the target kerosene from narrowing, and to prevent the deterioration of the combustibility of the target kerosene.

A tenth aspect of the present disclosure is an information processing device according to any one of the seventh to ninth aspects, in which the calculation unit determines the target blending ratio so that the difference between the a% distillation temperature and the (a+25)% distillation temperature of the target kerosene is 7°C or more. According to the tenth aspect, it is possible to prevent the distillation properties (distillation range) of the target kerosene from narrowing, and to prevent the deterioration of the combustibility of the target kerosene.

An eleventh aspect of the present disclosure is the information processing device according to any one of the seventh to tenth aspects, in which the calculation unit determines the target blending ratio so that the difference between the a% distillation temperature and the (a+20)% distillation temperature of the target kerosene is 6°C or more. According to the eleventh aspect, it is possible to prevent the distillation properties (distillation range) of the target kerosene from narrowing, and to prevent the deterioration of the combustibility of the target kerosene.

A twelfth aspect of the present disclosure is the information processing device according to any one of the seventh to eleventh aspects, wherein the plurality of parameters further includes a kinetic viscosity, and the calculation unit determines the target blending ratio so that the following conditions are satisfied for the kinetic viscosity v (30° C., mm ² /s) and density ρ (15° C., g/cm ³ ) of the target kerosene: v≦1.522, ρ≦0.8061, and ρ≦−0.0429v+0.08653. According to the twelfth aspect, a target blending ratio for producing a target kerosene with excellent long-term combustibility can be determined by using a constraint condition that limits the density and kinetic viscosity of the target kerosene to a predetermined range.

A thirteenth aspect of the present disclosure is the information processing device according to any one of the seventh to twelfth aspects, in which the plurality of parameters further includes a sulfur content, and the calculation unit determines the target blend ratio so that the sulfur content of the target kerosene is equal to or less than a predetermined reference value. According to the thirteenth aspect, the blend ratio is determined so as to satisfy the reference value of the sulfur content, thereby improving user convenience.

A fourteenth aspect of the present disclosure is the information processing device according to any one of the seventh to thirteenth aspects, in which the plurality of parameters further includes a calorific value, and the calculation unit determines the target blend ratio so that the calorific value of the target kerosene satisfies a predetermined condition. According to the fourteenth aspect, the blend ratio is determined so as to satisfy a condition related to the calorific value, thereby improving user convenience.

A fifteenth aspect of the present disclosure is the information processing device according to any one of the seventh to fourteenth aspects, in which the base oil data includes the respective costs of the plurality of base oils, and the calculation unit determines the target blend ratio so that the cost of the target kerosene satisfies a predetermined condition. According to the fifteenth aspect, the blend ratio is determined so as to satisfy a cost-related condition, thereby improving user convenience.

A sixteenth aspect of the present disclosure is the information processing device according to any one of the seventh to fifteenth aspects, in which the base oil data includes the inventory amounts of each of the multiple base oils, and the calculation unit determines the target blending ratio such that the usage amount of each of the multiple base oils used when producing the target kerosene at a predetermined production volume is equal to or less than the inventory amount. According to the sixteenth aspect, it is possible to utilize limited inventory of base oils, which can contribute to increased production of kerosene with excellent combustion properties.

In a seventeenth aspect of the present disclosure, the plurality of base stocks include a first base stock and a second base stock, and the first base stock is selected from a first group including straight-run kerosene, a kerosene fraction obtained by hydrocracking straight-run kerosene, a kerosene fraction obtained by hydrocracking straight-run heavy oil, a kerosene fraction obtained by hydrocracking catalytic cracking light oil at a low cracking rate, a kerosene fraction obtained by fractionating and desulfurizing catalytic cracking light oil, a desulfurized kerosene fraction derived from oil sands, and a residual oil obtained by distilling and separating and removing a fraction having a temperature of 160°C or higher and 190°C or lower from a kerosene fraction, and the second base stock is selected from a first group including norpara raffinate, a kerosene fraction obtained by hydrocracking catalytic cracking light oil at a high cracking rate, C 9 aromatic hydrocarbons, isooctene heavy, FT synthetic oil, kerosene fraction obtained by fractionating FT synthetic oil, kerosene fraction obtained by hydrogenating FT synthetic oil, kerosene fraction obtained by isomerizing FT synthetic oil, bio-derived oil, kerosene fraction obtained by fractionating bio-derived oil, kerosene fraction obtained by hydrogenating bio-derived oil, kerosene fraction obtained by isomerizing bio-derived oil, waste oil-derived reclaimed oil, kerosene fraction obtained by fractionating waste oil-derived reclaimed oil, kerosene fraction obtained by hydrogenating waste oil-derived reclaimed oil, and kerosene fraction obtained by isomerizing waste oil-derived reclaimed oil. According to the 17th aspect, by combining a first base oil having a relatively high naphthenobenzene content with a second base oil having a relatively low naphthenobenzene content, it is possible to determine the blending ratio of a target kerosene in which the naphthenobenzene content is equal to or less than a predetermined reference value. This allows for the effective use of the first base stock, which contains a relatively large amount of naphthenobenzene, and contributes to increased kerosene production.

In an eighteenth aspect of the present disclosure, the plurality of base stocks include a first base stock and a second base stock, the first base stock being selected from a first group including a kerosene fraction obtained by hydrocracking catalytic cracking diesel at a high cracking rate, C9 aromatic hydrocarbons, isooctene heavy, bio-derived oil, a kerosene fraction obtained by fractionating bio-derived oil, a kerosene fraction obtained by hydrogenating bio-derived oil, a kerosene fraction obtained by isomerizing bio-derived oil, a waste oil-derived recycled oil, a kerosene fraction obtained by fractionating waste oil-derived recycled oil, a kerosene fraction obtained by hydrogenating waste oil-derived recycled oil, and a kerosene fraction obtained by isomerizing waste oil-derived recycled oil, and the second base stock is is an information processing device according to any one of the seventh to sixteenth aspects, which is selected from a second group including a kerosene fraction obtained by hydrodesulfurizing straight-run kerosene, a kerosene fraction obtained by hydrocracking straight-run heavy oil, a kerosene fraction obtained by hydrocracking catalytic cracked light oil at a low cracking rate, a kerosene fraction obtained by fractionating and desulfurizing catalytic cracked light oil, a desulfurized kerosene fraction derived from oil sands, a residual oil obtained after extracting a fraction having a temperature of 160°C or higher and 190°C or lower from a kerosene fraction, norpara raffinate, FT synthetic oil, a kerosene fraction obtained by fractionating FT synthetic oil, a kerosene fraction obtained by hydrogenating FT synthetic oil, and a kerosene fraction obtained by isomerizing FT synthetic oil. According to the eighteenth aspect, by combining a first base oil having a relatively narrow distillation property (temperature range) with a second base oil having a relatively narrow distillation property (temperature range), it is possible to suppress the distillation property (distillation range) of the target kerosene from narrowing, and to suppress deterioration of the combustibility of the target kerosene. This allows the first base stock, which has a relatively narrow distillation property (temperature range), to be used effectively, contributing to increased kerosene production.

A 19th aspect of the present disclosure is the information processing device according to any one of the 7th to 18th aspects, in which the multiple base oils include non-petroleum-derived base oils. According to the 19th aspect, even when a non-petroleum-derived base oil is used, it is possible to determine a target blending ratio for producing a target kerosene in which the naphthenobenzene content is equal to or less than a predetermined reference value.

A twentieth aspect of the present disclosure is a prediction method comprising: acquiring model data relating to a correlation model showing a correlation between a plurality of parameters indicating the properties of each of a plurality of kerosenes, the plurality of parameters including at least density, distillation properties, and aromatic content, and the amount of naphthenobenzene contained in each of the plurality of kerosenes; acquiring numerical values of the plurality of parameters for a specific kerosene; and predicting the amount of naphthenobenzene contained in the specific kerosene using the model data and the numerical values of the plurality of parameters for the specific kerosene. According to the twentieth aspect, the amount of naphthenobenzene, which requires a lot of time and effort to analyze, can be predicted using parameters that can be obtained at the kerosene production site. In particular, by using at least density, distillation properties, and aromatic content as the plurality of parameters, the amount of naphthenobenzene contained in the specific kerosene can be predicted with high accuracy.

A twenty-first aspect of the present disclosure is a program that causes a computer to realize a function of acquiring model data relating to a correlation model that indicates a correlation between a plurality of parameters indicating the properties of each of a plurality of kerosenes, the plurality of parameters including at least density, distillation properties, and aromatic content, and the amount of naphthenobenzene contained in each of the plurality of kerosenes, a function of acquiring the numerical values of the plurality of parameters of a specific kerosene, and a function of predicting the amount of naphthenobenzene contained in the specific kerosene using the model data and the numerical values of the plurality of parameters of the specific kerosene. According to the twenty-first aspect, the amount of naphthenobenzene, which is time-consuming to analyze, can be predicted using parameters that can be acquired at the kerosene production site. In particular, by using at least density, distillation properties, and aromatic content as the plurality of parameters, the amount of naphthenobenzene contained in the specific kerosene can be predicted with high accuracy.

The 22nd aspect of the present disclosure is an information processing device including: a storage unit that stores model data relating to a correlation model that shows a correlation between a plurality of parameters indicating the properties of each of a plurality of kerosenes, the plurality of parameters including at least density, distillation properties, and aromatic content, and the amount of naphthenobenzene contained in each of the plurality of kerosenes, and stores base oil data including the numerical values of the plurality of parameters for each of a plurality of base oils used in the production of a target kerosene; and a calculation unit that uses the model data and the base oil data to determine a target blending ratio of the plurality of base oils for producing the target kerosene so that the amount of naphthenobenzene contained in the target kerosene is equal to or less than a predetermined reference value. According to the 22nd aspect, it is possible to determine a target blending ratio for producing a target kerosene in which the amount of naphthenobenzene is equal to or less than a predetermined reference value. In particular, since it is not necessary to analyze the amount of naphthenobenzene contained in each of a plurality of base oils used in the production of the target kerosene, the effort of analysis can be significantly reduced.

Any combination of the configurations according to the above-mentioned embodiments or aspects is also useful as an embodiment of the present disclosure. A new embodiment resulting from the combination has the effects of each of the combined examples and modifications. It will also be understood by those skilled in the art that the functions to be performed by each of the constituent elements described in the claims can be realized by each of the components shown in the examples and modifications alone or in combination with each other.

10: information processing device, 12: acquisition unit, 14: memory unit, 16: calculation unit, 20: base oil data, 22: kerosene data, 24: model data, 30: model generation unit, 32: prediction unit, 34: blending determination unit.

Claims (22)

A storage unit that stores model data relating to a correlation model that shows a correlation between a plurality of parameters indicating properties of each of a plurality of kerosenes, the plurality of parameters including at least density, distillation properties, and aromatic content, and the amount of naphthenobenzene contained in each of the plurality of kerosenes;
An acquisition unit that acquires numerical values of the plurality of parameters of a specific kerosene;
and a calculation unit that predicts an amount of naphthenobenzene contained in the specific kerosene using the model data and numerical values of the plurality of parameters of the specific kerosene. The information processing device according to claim 1, wherein the distillation properties include at least one of a 50% distillation temperature, a 70% distillation temperature, and a 90% distillation temperature. The information processing device according to claim 1, wherein the distillation properties include at least one of a distillation percentage at 210°C or higher, a distillation percentage at 235°C or higher, and a distillation percentage at 240°C or higher. The information processing device according to claim 1, wherein the aromatic content includes a total aromatic content and a polycyclic aromatic content. The information processing device according to claim 1, wherein the plurality of parameters further includes kinetic viscosity. The information processing device according to claim 1, wherein the model data is generated by machine learning using training data in which the numerical values of the parameters of the multiple kerosene oils are input and the amount of naphthenobenzene contained in the multiple kerosene oils is output. The memory unit further stores base oil data including numerical values of the plurality of parameters for each of a plurality of base oils used in the production of the target kerosene;
7. The information processing device according to claim 1, wherein the calculation unit determines a target blending ratio of the plurality of base oils for producing the target kerosene, using the model data and the base oil data, so that an amount of naphthenobenzene contained in the target kerosene is equal to or less than a predetermined reference value. The information processing device according to claim 7, wherein the calculation unit determines the target blending ratio so that the amount of naphthenobenzene contained in the target kerosene is 8.9% by volume or less. The information processing device according to claim 7, wherein the calculation unit determines the target blending ratio so that the difference between the 10% distillation temperature and the 50% distillation temperature of the target kerosene is greater than 10°C. The information processing device according to claim 7, wherein the calculation unit determines the target blending ratio so that the difference between the a% distillation temperature of the target kerosene and the (a+25)% distillation temperature is 7°C or more. The information processing device according to claim 7, wherein the calculation unit determines the target blending ratio so that the difference between the a% distillation temperature of the target kerosene and the (a+20)% distillation temperature is 6°C or more. The plurality of parameters further comprises kinematic viscosity;
The information processing device according to claim 7, wherein the calculation unit determines the target blending ratio so that the conditions of ν≦1.522, ρ≦0.8061, and ρ≦−0.0429ν+0.08653 are satisfied for the kinetic viscosity ν (30°C, ^mm 2 /s) and density ρ (15°C, g/cm ³ ) of the target kerosene. The plurality of parameters further comprises sulfur content;
The information processing device according to claim 7 , wherein the calculation unit determines the target blending ratio so that a sulfur content of the target kerosene is equal to or less than a predetermined reference value. The plurality of parameters further comprises a heat value;
The information processing device according to claim 7 , wherein the calculation unit determines the target blending ratio so that the target heating value of the kerosene satisfies a predetermined condition. the base stock data includes a cost of each of the plurality of base stocks;
The information processing device according to claim 7 , wherein the calculation unit determines the target blend ratio so that the target kerosene cost satisfies a predetermined condition. the base stock data includes an inventory amount of each of the plurality of base stocks;
The information processing device according to claim 7 , wherein the calculation unit determines the target blending ratio so that the usage amount of each of the plurality of base oils used when producing the target kerosene at a predetermined production volume is equal to or less than the inventory amount. The plurality of base oils include a first base oil and a second base oil,
The first base oil is selected from a first group including straight-run kerosene, a kerosene fraction obtained by hydrocracking straight-run kerosene, a kerosene fraction obtained by hydrocracking straight-run heavy oil, a kerosene fraction obtained by hydrocracking catalytic cracking gas oil at a low cracking rate, a kerosene fraction obtained by fractionating and desulfurizing catalytic cracking gas oil, a desulfurized kerosene fraction derived from oil sands, and a residual oil obtained by distilling and separating a fraction having a temperature of 160° C. or higher and 190° C. or lower from a kerosene fraction,
The information processing device according to claim 7, wherein the second base oil is selected from a second group including norpara raffinate, a kerosene fraction obtained by hydrocracking catalytic cracking diesel at a high cracking rate, C9 aromatic hydrocarbons, isooctene heavy, FT synthetic oil, a kerosene fraction obtained by fractionating FT synthetic oil, a kerosene fraction obtained by hydrogenating FT synthetic oil, a kerosene fraction obtained by isomerizing FT synthetic oil, a bio-derived oil, a kerosene fraction obtained by fractionating bio-derived oil, a kerosene fraction obtained by hydrogenating bio-derived oil, a kerosene fraction obtained by isomerizing bio-derived oil, a waste oil-derived reclaimed oil, a kerosene fraction obtained by fractionating waste oil-derived reclaimed oil, a kerosene fraction obtained by hydrogenating waste oil-derived reclaimed oil, and a kerosene fraction obtained by isomerizing waste oil-derived reclaimed oil. The plurality of base oils include a first base oil and a second base oil,
The first base stock is selected from a first group including a kerosene fraction obtained by hydrocracking catalytic cracking gas oil at a high cracking rate, a C9 aromatic hydrocarbon, an isooctene heavy, a bio-derived oil, a kerosene fraction obtained by fractionating a bio-derived oil, a kerosene fraction obtained by hydrogenating a bio-derived oil, a kerosene fraction obtained by isomerizing a bio-derived oil, a waste oil-derived recycled oil, a kerosene fraction obtained by fractionating a waste oil-derived recycled oil, a kerosene fraction obtained by hydrogenating a waste oil-derived recycled oil, and a kerosene fraction obtained by isomerizing a waste oil-derived recycled oil;
8. The information processing device according to claim 7, wherein the second base oil is selected from a second group including a kerosene fraction obtained by hydrocracking straight-run kerosene, a kerosene fraction obtained by hydrocracking straight-run heavy oil, a kerosene fraction obtained by hydrocracking catalytically cracked light oil at a low cracking rate, a kerosene fraction obtained by fractionating and desulfurizing catalytically cracked light oil, a desulfurized kerosene fraction derived from oil sands, a residual oil obtained after extracting a fraction having a temperature of 160°C or higher and 190°C or lower from a kerosene fraction, nor-para-raffinate, FT synthetic oil, a kerosene fraction obtained by fractionating FT synthetic oil, a kerosene fraction obtained by hydrogenating FT synthetic oil, and a kerosene fraction obtained by isomerizing FT synthetic oil. The information processing device according to claim 7, wherein the plurality of base oils include non-petroleum-derived base oils. Acquiring model data relating to a correlation model showing a correlation between a plurality of parameters indicating properties of each of a plurality of kerosenes, the plurality of parameters including at least density, distillation properties, and aromatic content, and an amount of naphthenobenzene contained in each of the plurality of kerosenes;
Obtaining values of the plurality of parameters for a particular kerosene;
predicting an amount of naphthenobenzene contained in the specific kerosene using the model data and the numerical values of the plurality of parameters of the specific kerosene. A function of acquiring model data relating to a correlation model showing a correlation between a plurality of parameters indicating properties of each of a plurality of kerosene oils, the plurality of parameters including at least density, distillation properties, and aromatic content, and the amount of naphthenobenzene contained in each of the plurality of kerosene oils;
Obtaining the numerical values of the plurality of parameters of a particular kerosene;
and a function of predicting an amount of naphthenobenzene contained in the specific kerosene, using the model data and the numerical values of the plurality of parameters of the specific kerosene. A storage unit stores model data relating to a correlation model showing a correlation between a plurality of parameters indicating properties of each of a plurality of kerosenes, the plurality of parameters including at least density, distillation properties, and aromatic content, and the amount of naphthenobenzene contained in each of the plurality of kerosenes, and stores base oil data including numerical values of the plurality of parameters for each of a plurality of base oils used in the production of a target kerosene;
and a calculation unit that uses the model data and the base oil data to determine a target blending ratio of the multiple base oils for producing the target kerosene so that the amount of naphthenobenzene contained in the target kerosene is equal to or less than a predetermined reference value.

Images