Method for predicting physical and chemical properties of crude oil

Abstract

The invention relates to a method for predicting physical and chemical properties of crude oil, and the method comprises the following steps: obtaining library sample spectrums of a plurality of crude oil samples, and obtaining respective standard data of a first physical and chemical property and a second physical and chemical property; taking the first n main factors of the first characteristic spectrum and the standard data of the first physicochemical property of the crude oil sample as a first variable, taking the standard data of the second physicochemical property of the crude oil sample as a second variable, and performing partial least square regression analysis to obtain a correction model; collecting a to-be-detected sample spectrum of a to-be-detected crude oil sample, and obtaining detection data of a first physicochemical property of the to-be-detected crude oil sample; substituting the first n main factors of the second characteristic spectrum and the detection data of the first physicochemical property of the crude oil sample to be detected into the correction model to obtain the prediction data of the second physicochemical property of the crude oil sample to be detected. The method disclosed by the invention has a relatively wide range of physicochemical property prediction capability and better prediction precision.

Description

Technical Field

[0001] The present disclosure relates to a method for predicting the physicochemical properties of crude oil. Background Art

[0002] Crude oil evaluation plays a very important role in crude oil extraction, crude oil trade and crude oil processing. Rapid detection of crude oil properties is of great significance for optimizing feed, blending and improving efficiency. Existing research on crude oil rapid evaluation technology is based on a variety of modern instrumental analysis methods, including chromatography, mass spectrometry, nuclear magnetic resonance and infrared / near infrared spectroscopy. Near-infrared spectroscopy analysis technology has the advantages of rapid and non-destructive analysis. It is used to determine the properties of crude oil and is a widespread and in-depth test project in the petrochemical field. There is a good linear relationship between the absorbance of the near-infrared spectrum and the physicochemical properties of crude oil. A more practical analysis model can be established through the partial least squares method. However, the evaluation results of these analysis models are often less accurate. Summary of the invention

[0003] The purpose of the present disclosure is to provide a method for predicting the physicochemical properties of crude oil to improve the prediction accuracy.

[0004] In order to achieve the above object, the present disclosure provides a method for predicting the physicochemical properties of crude oil, the method comprising the following steps:

[0005] Obtaining library sample spectra of a plurality of crude oil samples, and obtaining standard data of a first physicochemical property and a second physicochemical property of each crude oil sample; wherein the first physicochemical property and the second physicochemical property respectively include at least one physicochemical property of the crude oil;

[0006] Obtaining the absorbance of a characteristic spectrum region in each of the library sample spectra to obtain a first characteristic spectrum, performing partial least squares regression analysis on the first characteristic spectrum to obtain first n principal factors of the first characteristic spectrum;

[0007] The first n principal factors of the first characteristic spectrum and the standard data of the first physicochemical property of the crude oil sample are used as the first variable, the standard data of the second physicochemical property of the crude oil sample is used as the second variable, and the first variable and the second variable are subjected to partial least squares regression analysis to obtain a calibration model;

[0008] Collecting a sample spectrum of the crude oil sample to be tested, and obtaining detection data of a first physicochemical property of the crude oil sample to be tested;

[0009] Obtaining the absorbance of a characteristic spectrum region of the spectrum of the sample to be tested to obtain a second characteristic spectrum, performing partial least squares regression analysis on the second characteristic spectrum to obtain the first n principal factors of the second characteristic spectrum;

[0010] Substitute the first n principal factors of the second characteristic spectrum and the detection data of the first physicochemical property of the crude oil sample to be tested into the calibration model to obtain the prediction data of the second physicochemical property of the crude oil sample to be tested.

[0011] Optionally, the characteristic spectrum range is 4000-4500cm -1 and 5500~6000cm -1 .

[0012] Optionally, n is an integer selected from 1-15.

[0013] Optionally, the first physicochemical property and the second physicochemical property are each independently selected from one or more of true boiling point distillation cut-off yield, density, acid value, residual carbon value, sulfur content, nitrogen content, wax content and colloid asphaltene content, and the first physicochemical property is different from the second physicochemical property.

[0014] Optionally, the first physicochemical property is the true boiling point distillation cut-off yield, and the second physicochemical property is selected from one or more of density, acid value, residual carbon value, sulfur content, nitrogen content, wax content and colloid asphaltene content.

[0015] Optionally, the density is tested according to standard method SH / T0604-2000;

[0016] The acid value is tested according to the standard method GB / T 7304;

[0017] The carbon residue value is tested according to the standard method GB / T 17144;

[0018] The sulfur content is tested according to the standard method GB / T17040;

[0019] The nitrogen content is tested according to the standard method GB / T 17674;

[0020] The wax content is tested according to standard method SY / T 0537;

[0021] The colloid asphaltene content is tested according to the standard method SY / T 7550;

[0022] The actual boiling point distillation cut yield is tested according to the standard method GB / T 17280.

[0023] Optionally, the method further comprises:

[0024] Multiple predictions are performed based on the first n principal factors of the second characteristic spectrum, the detection data of the first physicochemical property of the crude oil sample to be tested, and the calibration model, and the average value of the multiple prediction results is taken as the prediction data of the second physicochemical property of the crude oil sample to be tested.

[0025] Optionally, the method comprises:

[0026] Performing second-order differential processing on each library sample spectrum to obtain a first differential spectrum;

[0027] The first differential absorbance of a characteristic spectrum region in each of the first differential spectra is acquired to obtain the first characteristic spectrum.

[0028] Optionally, the method comprises:

[0029] Performing second-order differential processing on the spectrum of the sample to be tested to obtain a second differential spectrum;

[0030] A second differential absorbance of a characteristic spectral region of the second differential spectrum is acquired to obtain the second characteristic spectrum.

[0031] Optionally, the window width of the second-order differential processing is 13 to 25.

[0032] Through the above technical solution, the present invention combines the library sample spectrum of the crude oil sample with the first physicochemical property and uses the partial least squares method to establish a linear correction model for the second physicochemical property, which has a wider range of physicochemical property prediction capabilities and better prediction accuracy.

[0033] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present disclosure but do not constitute a limitation of the present disclosure. In the accompanying drawings:

[0035] Figure 1 It is a schematic flow chart of a method for predicting the physicochemical properties of crude oil provided in one embodiment of the present disclosure. DETAILED DESCRIPTION

[0036] The specific implementation of the present disclosure is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described herein is only used to illustrate and explain the present disclosure, and is not used to limit the present disclosure.

[0037] The present disclosure provides a method for predicting the physicochemical properties of crude oil, such as Figure 1 As shown, the following steps S101 to S106 are included:

[0038] S101, obtaining library sample spectra of a plurality of crude oil samples, and obtaining respective standard data of a first physicochemical property and a second physicochemical property of each of the crude oil samples; wherein the first physicochemical property and the second physicochemical property respectively include at least one physicochemical property of the crude oil;

[0039] S102, obtaining the absorbance of a characteristic spectrum region in each of the library sample spectra to obtain a first characteristic spectrum, performing partial least squares regression analysis on the first characteristic spectrum to obtain first n principal factors of the first characteristic spectrum;

[0040] S103, using the first n principal factors of the first characteristic spectrum and the standard data of the first physicochemical property of the crude oil sample as the first variable, using the standard data of the second physicochemical property of the crude oil sample as the second variable, and performing partial least squares regression analysis on the first variable and the second variable to obtain a calibration model;

[0041] S104, collecting a sample spectrum of the crude oil sample to be tested, and obtaining detection data of a first physicochemical property of the crude oil sample to be tested;

[0042] S105, obtaining the absorbance of the characteristic spectrum region of the spectrum of the sample to be tested to obtain a second characteristic spectrum, performing partial least squares regression analysis on the second characteristic spectrum to obtain the first n principal factors of the second characteristic spectrum;

[0043] S106, substituting the first n principal factors of the second characteristic spectrum and the detection data of the first physicochemical property of the crude oil sample to be tested into the calibration model to obtain prediction data of the second physicochemical property of the crude oil sample to be tested.

[0044] In the present disclosure, the library sample spectrum and the sample spectrum to be tested are both near-infrared spectra. By combining the library sample spectrum of the crude oil sample with the first physicochemical property, a linear correction model is established for the second physicochemical property using the partial least squares method, which has a wider range of physicochemical property prediction capabilities and better prediction accuracy.

[0045] In the present disclosure, partial least squares regression (PLS) analysis can be performed using conventional selection methods and software in the art, such as nonlinear iterative partial least squares algorithm (NIPALS).

[0046] In one embodiment, n is an integer selected from 1 to 15. In this embodiment, selecting an appropriate number of principal factors n is beneficial to simplifying the amount of calculation in the process of building the correction model and further improving the accuracy of the prediction results of the physicochemical properties.

[0047] In a specific embodiment, the near infrared spectroscopy detection conditions include: the scanning range is 4000-10000cm -1 In the present disclosure, the resolution can be selected according to the calibration model construction accuracy or instrument accuracy, for example, the resolution is 4 cm -1 、8cm -1 wait.

[0048] In one embodiment, the characteristic spectrum range is 4000-4500 cm -1 and 5500~6000cm -1 In this embodiment, the absorbance within the appropriate characteristic spectrum range is selected to construct the model, which can improve the correlation between absorbance and physicochemical properties, simplify the amount of calculation in the correction model construction process, and further improve the accuracy of the physicochemical property prediction results.

[0049] In the present disclosure, the first physicochemical property and the second physicochemical property refer to the physical and / or chemical properties of the crude oil sample, and the first physicochemical property is different from the second physicochemical property, wherein the second physicochemical property is the physicochemical property that needs to be predicted.

[0050] The number of types of the first physicochemical property and the second physicochemical property can be selected as needed. For example, the first physicochemical property may include 1 to 3 types of crude oil physicochemical properties, and the second physicochemical property includes the physicochemical properties of crude oil excluding the first physicochemical property. Selecting an appropriate number of first physicochemical properties and second physicochemical properties is conducive to simplifying the amount of calculation in the process of building the correction model, and further improving the accuracy of the physicochemical property prediction results.

[0051] The types of the first physicochemical property and the second physicochemical property can be selected as needed. For example, the first physicochemical property may include physicochemical properties that are easier to predict and / or secondary physicochemical properties, and the second physicochemical property may include physicochemical properties that are more difficult to predict and / or key physicochemical properties. Specifically, the first physicochemical property and the second physicochemical property are each independently selected from one or more of true boiling point distillation cut yield, density, acid value, residual carbon value, sulfur content, nitrogen content, wax content and colloid asphaltene content, and the first physicochemical property is different from the second physicochemical property.

[0052] In one embodiment, the first physicochemical property is the actual boiling point distillation cut-off yield; the second physicochemical property is selected from one or more of density, acid value, residual carbon value, sulfur content, nitrogen content, wax content, colloid, and asphaltene content. This embodiment is conducive to further improving the accuracy of the prediction result of the second physicochemical property.

[0053] Furthermore, each physicochemical property is measured by standard methods. Specifically, the density is measured according to the standard method SH / T0604-2000; the acid value is measured according to the standard method GB / T 7304; the carbon residue value is measured according to the standard method GB / T 17144; the sulfur content is measured according to the standard method GB / T17040; the nitrogen content is measured according to the standard method GB / T 17674; the wax content is measured according to the standard method SY / T 0537; the colloid asphaltene content is measured according to the standard method SY / T 7550; the true boiling point distillation cut yield is measured according to the standard method GB / T 17280.

[0054] In one embodiment, the method further comprises: performing multiple predictions based on the first n principal factors of the second characteristic spectrum, the detection data of the first physicochemical property of the crude oil sample to be tested, and the calibration model, and taking the average value of the multiple prediction results as the prediction data of the second physicochemical property of the crude oil sample to be tested. This is conducive to avoiding errors and improving the accuracy of the prediction results.

[0055] In one embodiment, the method further includes: performing a second-order differential process on each of the library sample spectra to obtain a first differential spectrum; obtaining a first differential absorbance of a characteristic spectrum region in each of the first differential spectra to obtain the first characteristic spectrum. Further, the method further includes: performing a second-order differential process on the spectrum of the sample to be tested to obtain a second differential spectrum; obtaining a second differential absorbance of a characteristic spectrum region of the second differential spectrum to obtain the second characteristic spectrum. By performing a second-order differential process on the library sample spectrum and / or the spectrum of the sample to be tested, the interference of the baseline offset, drift and background of the spectrum can be eliminated.

[0056] In a specific implementation manner, the window width of the second-order differential processing is 13 to 25. In the present disclosure, the second-order differential processing method can be derived using Savitzky-Golay (SG).

[0057] In the present disclosure, the crude oil samples can be the same type of crude oil or different types of crude oil. The present disclosure has no particular restrictions on the type of crude oil samples. In a specific embodiment, the number of the crude oil samples is at least 200 to further improve the accuracy and reliability of the calibration model established.

[0058] In the present disclosure, crude oil samples are divided into calibration set samples and test set samples, wherein the calibration set samples are used for model construction, and the test set samples are used to verify the predicted values ​​and actual values ​​of the constructed model. In the present disclosure, a large number of calibration set samples can be used, and they have good representativeness, that is, the physicochemical properties of the calibration set samples can cover the physicochemical properties of all predicted crude oil samples, so that the established model has good generalization ability. Preferably, the number of calibration set samples accounts for 2 / 3 to 3 / 4 of the total number of samples, and the remaining 1 / 3 to 1 / 4 are used as test set samples. Preferably, the distribution laws of the calibration set and test set samples used are approximately the same.

[0059] The present disclosure is further described in detail below through examples, but the present disclosure is not limited thereto.

[0060] In the following embodiments and comparative examples, the test conditions for near infrared spectroscopy tests on calibration set samples and test set samples are the same.

[0061] The instrument for near infrared spectroscopy was a Fourier transform near infrared spectrometer produced by Thermo, model Antaris II, and the spectral acquisition conditions were: 0.5 mm cuvette, resolution 8 cm -1 , the acquisition wave number range is 4000~10000cm -1 , cumulative scanning times 64 times, transmission measurement method.

[0062] The density is tested according to the standard method SH / T0604-2000; the acid value is tested according to the standard method GB / T 7304; the residual carbon value is tested according to the standard method GB / T 17144; the sulfur content is tested according to the standard method GB / T17040; the nitrogen content is tested according to the standard method GB / T17674; the wax content is tested according to the standard method SY / T 0537; the colloid asphaltene content is tested according to the standard method SY / T 7550; the true boiling point distillation cut yield is tested according to the standard method GB / T 17280.

[0063] Example 1

[0064] (1) Sample collection and determination of physicochemical properties

[0065] A total of 300 typical crude oil samples were collected, and the physicochemical properties data were tested using standard methods. The true boiling point distillation cut yield was selected as the first physicochemical property, and the density, acid value, residual carbon value, sulfur content, nitrogen content, wax content, and colloid asphaltene content were selected as the second physicochemical property. 280 representative crude oil samples were selected to form a calibration set, and the remaining 20 samples were used as test set samples to verify the accuracy of the model. The calibration set samples tested the first and second physicochemical properties as standard data, and the test set samples only tested the first physicochemical property data.

[0066] (2) Establishing a calibration model using calibration set samples

[0067] All calibration set samples are subjected to near infrared spectroscopy detection to obtain first near infrared spectra of all calibration set samples, and the first near infrared spectrum of each calibration set sample is subjected to second-order differential processing with a window width of 13 to 25 to obtain a first differential spectrum.

[0068] Intercept each first differential spectrum and select 4000~4500cm -1 、5500~6000cm -1 The absorbance in the spectral region constitutes a first characteristic spectrum, and the first 10 main factors thereof are obtained by partial least squares (PLS) method, and the first 10 main factors are combined with the standard data of the actual boiling point distillation cut yield of the first physicochemical property in step (1) as the first variable X, and the standard data of the second physicochemical property measured in step (1) constitute the second variable Y. The partial least squares (PLS) method is used to establish a calibration model of the X matrix and the Y matrix.

[0069] (3) Verify the accuracy of the calibration model

[0070] A near infrared spectrum test was performed on each sample in the test set to obtain a second near infrared spectrum. After a second-order differential process with a window width of 13 to 25, a second differential spectrum was obtained. The wave number of the second differential spectrum was intercepted to be 4000 to 4500 cm -1 、5500~6000cm -1 The absorbance of the spectral region constitutes the second characteristic spectrum, and the first 10 main factors thereof are obtained by partial least squares (PLS) method, and the first 10 main factors are combined with the detection data of the first physicochemical property actual boiling point distillation cut yield in step (1) and substituted into the calibration model to obtain the predicted data of the second physicochemical property of the test set samples. The comparison results of the density, residual carbon value, and wax content predicted values ​​of the 20 test set samples with the measured values ​​by the standard method are shown in Table 1, and the comparison results of the predicted values ​​of the remaining second physicochemical properties of the first sample with the measured values ​​by the standard method are shown in Table 2.

[0071] The calculation formula of relevant statistical parameters is as follows:

[0072]

[0073] in, is the predicted value of the i-th sample, y i (i=1,2,…,n) is the true value of the i-th sample, and n is the number of samples in the test set.

[0074] Deviation = measured value - predicted value.

[0075] Table 1

[0076]

[0077] Table 2

[0078]

[0079]

[0080] Example 2

[0081] The physicochemical properties of crude oil are predicted according to the method of Example 1, except that in step (1), acid value, sulfur content, and nitrogen content are selected as the first physicochemical property, and density, carbon residue value, and wax content are selected as the second physicochemical property.

[0082] The comparison results of the density, carbon residue value and wax content predicted values ​​of the test set samples with the values ​​determined by the standard method are shown in Table 3.

[0083] Table 3

[0084]

[0085] Comparative Example 1

[0086] The physicochemical properties of crude oil are predicted according to the method of Example 1, except that the near infrared spectrum is directly associated with the physicochemical properties of crude oil to establish a PLS calibration model. The specific steps are as follows:

[0087] (1) Sample collection and determination of physicochemical properties

[0088] A total of 300 typical crude oil samples were collected and the physicochemical properties data, including density, acid value, residual carbon, sulfur content, nitrogen content, wax content, colloid asphaltene content, and true boiling point distillation cut-off yield, were tested using standard methods. 280 representative crude oil samples were selected to form the calibration set, and the remaining 20 samples were used as test set samples to verify the accuracy of the model.

[0089] (2) Establishing a calibration model using calibration set samples

[0090] Perform near infrared spectroscopy on all calibration set samples to obtain the first near infrared spectra of all calibration set samples. Perform second-order differential processing on the first near infrared spectrum of each calibration set sample with a window width of 13 to 25 to obtain the first differential spectrum. Select 4000 to 4500 cm from each first differential spectrum. -1 、5500~6000cm -1 The absorbance in the spectral region constitutes the first characteristic spectrum, which is used as the first variable X. The standard data of all physicochemical properties measured in step (1) constitute the second variable Y. The calibration model of the X matrix and the Y matrix is ​​established by partial least squares (PLS).

[0091] (3) Verify the accuracy of the calibration model

[0092] A near infrared spectrum test was performed on each sample in the test set to obtain a second near infrared spectrum. After a second-order differential process with a window width of 13 to 25, a second differential spectrum was obtained. The wave number of the second differential spectrum was intercepted to be 4000 to 4500 cm -1 、5500~6000cm -1 The absorbance of the spectral region constitutes the second characteristic spectrum, which is substituted into the calibration model to obtain the predicted data of the physicochemical properties of the test set samples. The comparison results of the density, carbon residue value, and wax content predicted values ​​of the 20 test set samples with the values ​​measured by the standard method are shown in Table 4, and the comparison results of the predicted values ​​of the remaining physicochemical properties of the first sample with the values ​​measured by the standard method are shown in Table 5.

[0093] Table 4

[0094]

[0095]

[0096] Table 5

[0097] Measured value Predicted value deviation Acid value, mgKOH / g 0.67 0.97 -0.30 Sulfur content, w% 0.55 0.50 0.05 Nitrogen content, w% 0.23 0.19 0.04 Gum content, w% 9.8 9.3 0.5 Asphaltene content, w% 0.9 1.6 -0.7

[0098] According to the relevant data of the above embodiments and comparative examples, the method provided by the present invention can obtain a lower standard deviation of the prediction of physicochemical properties, which indicates that the method provided by the present invention can effectively improve the prediction accuracy of the physicochemical properties of crude oil.

[0099] The preferred embodiments of the present disclosure are described in detail above in conjunction with the accompanying drawings; however, the present disclosure is not limited to the specific details in the above embodiments. Within the technical concept of the present disclosure, a variety of simple modifications can be made to the technical solution of the present disclosure, and these simple modifications all fall within the protection scope of the present disclosure.

[0100] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present disclosure will not further describe various possible combinations.

[0101] In addition, various embodiments of the present disclosure may be arbitrarily combined, and as long as they do not violate the concept of the present disclosure, they should also be regarded as the contents disclosed by the present disclosure.

Claims

1. A method for predicting the physicochemical properties of crude oil, characterized in that: The method comprises the following steps: Obtaining library sample spectra of a plurality of crude oil samples, and obtaining standard data of a first physicochemical property and a second physicochemical property of each crude oil sample; wherein the first physicochemical property and the second physicochemical property respectively include at least one physicochemical property of the crude oil; Obtaining the absorbance of a characteristic spectrum region in each of the library sample spectra to obtain a first characteristic spectrum, performing partial least squares regression analysis on the first characteristic spectrum to obtain first n principal factors of the first characteristic spectrum; The first n principal factors of the first characteristic spectrum and the standard data of the first physicochemical property of the crude oil sample are used as the first variable, the standard data of the second physicochemical property of the crude oil sample is used as the second variable, and the first variable and the second variable are subjected to partial least squares regression analysis to obtain a calibration model; Collecting a sample spectrum of the crude oil sample to be tested, and obtaining detection data of a first physicochemical property of the crude oil sample to be tested; Obtaining the absorbance of a characteristic spectrum region of the spectrum of the sample to be tested to obtain a second characteristic spectrum, performing partial least squares regression analysis on the second characteristic spectrum to obtain the first n principal factors of the second characteristic spectrum; Substitute the first n principal factors of the second characteristic spectrum and the detection data of the first physicochemical property of the crude oil sample to be tested into the calibration model to obtain the prediction data of the second physicochemical property of the crude oil sample to be tested.

2. The method according to claim 1, wherein: The characteristic spectrum range is 4000-4500 cm -1 and 5500~6000cm -1 .

3. The method according to claim 1, wherein: n is an integer selected from 1-15.

4. The method according to claim 1, wherein: The first physicochemical property and the second physicochemical property are each independently selected from one or more of true boiling point distillation cut yield, density, acid value, residual carbon value, sulfur content, nitrogen content, wax content and colloid asphaltene content, and the first physicochemical property is different from the second physicochemical property.

5. The method according to claim 1 or 4, wherein: The first physicochemical property is the actual boiling point distillation cut-off yield, and the second physicochemical property is selected from one or more of density, acid value, residual carbon value, sulfur content, nitrogen content, wax content and colloid asphaltene content.

6. The method according to claim 4, wherein: The density is tested according to standard method SH / T0604-2000; The acid value is tested according to the standard method GB / T 7304; The carbon residue value is tested according to the standard method GB / T 17144; The sulfur content is tested according to the standard method GB / T17040; The nitrogen content is tested according to the standard method GB / T 17674; The wax content is tested according to standard method SY / T 0537; The colloid asphaltene content is tested according to the standard method SY / T 7550; The actual boiling point distillation cut yield is tested according to the standard method GB / T 17280.

7. The method according to claim 1, wherein: The method further includes: Multiple predictions are performed based on the first n principal factors of the second characteristic spectrum, the detection data of the first physicochemical property of the crude oil sample to be tested, and the calibration model, and the average value of the multiple prediction results is taken as the prediction data of the second physicochemical property of the crude oil sample to be tested.

8. The method according to claim 1, wherein: The method includes: Performing second-order differential processing on each library sample spectrum to obtain a first differential spectrum; The first differential absorbance of a characteristic spectrum region in each of the first differential spectra is acquired to obtain the first characteristic spectrum.

9. The method according to claim 8, wherein: The method includes: Performing second-order differential processing on the spectrum of the sample to be tested to obtain a second differential spectrum; The second differential absorbance of the characteristic spectrum region of the second differential spectrum is acquired to obtain the second characteristic spectrum.

10. The method according to claim 8 or 9, wherein: The window width of the second-order differential processing is 13-25.

Images