Rapid screening method for crude oil / fuel oil commodity attribute identification

Abstract

The invention relates to a rapid screening method for identifying the commodity attributes of crude oil / fuel oil, which comprises the following steps: firstly analyzing the content of metal elements, especially nickel, in an oil product sample by using an X-ray fluorescence spectrometer, and then analyzing the molecular composition of the oil product, especially the chain length and distribution condition of hydrocarbon molecules, by using a micro-fluidic chip technology. According to the comprehensive analysis result of the metal element content and the molecular composition, the oil product type is classified and judged by combining a preset linear regression model, and whether the oil product is crude oil or fuel oil is judged. The method can quickly and accurately complete field detection, is suitable for industries such as oil fields, oil refineries and petroleum trade, and can output analysis results in real time through portable equipment. Compared with a traditional oil product identification method, the method has the advantages of being easy and convenient to operate, high in detection speed, high in precision, low in cost and the like, and is particularly suitable for application scenes needing on-site rapid detection.

Description

Technical Field

[0001] The invention relates to petrochemical analysis technology, in particular to a rapid screening method for identifying crude oil / fuel oil commodity attributes. Background Art

[0002] Crude oil and fuel oil are the two most common oil products in the oil industry. They have significant differences in production and use, mainly reflected in chemical composition, physical properties, uses, and impacts on the environment and equipment. Crude oil is a natural oil collected from underground through oil wells, and usually needs to go through a series of refining processes to be converted into various petroleum products, such as gasoline, diesel, kerosene, etc. Fuel oil is mainly heavy oil separated during the refining process, and is usually used in industrial combustion, ship engines, boilers and other equipment. Due to the differences in composition and use between crude oil and fuel oil, accurately identifying these two oil products has important application value, especially in the fields of oil processing, oil product trade, environmental protection, etc.

[0003] Existing methods for identifying crude oil and fuel oil mainly include physical and chemical analysis, instrument detection and chemical reaction methods. Physical and chemical analysis determines the type of oil by measuring characteristic parameters such as viscosity, density and flash point of oil, but this method has disadvantages such as low accuracy, cumbersome operation and the need for special equipment. Instrument detection methods such as mass spectrometry (MS), atomic absorption spectroscopy (AAS), gas chromatography (GC), etc., can accurately analyze the elemental and molecular composition of oil and are common high-precision oil identification methods. However, these methods often require complex laboratory equipment, the analysis process takes a long time, and the cost is high, which is not suitable for rapid detection on site.

[0004] In recent years, with the development of portable analytical technology, some innovative methods suitable for on-site detection have emerged. For example, portable X-ray fluorescence spectrometry (XRF) analyzers can quickly detect metal elements in samples and have been widely used in metal minerals, soil and water quality analysis and other fields (patent number: CN107309798A). This technology can effectively distinguish crude oil from fuel oil by analyzing the changes in the content of metal elements such as nickel and vanadium. Other patents, such as CN104748596A, propose oil identification methods based on chemical reactions and molecular sieving technologies, which process oil products through chemical reactions to identify their types. Although these existing methods have improved the efficiency of oil identification to a certain extent, there are still some problems, such as the need for special reagents, cumbersome operating steps, or can only be used in specific experimental environments, which limits their widespread promotion in practical applications.

[0005] Therefore, a new technical solution is urgently needed to combine the existing portable analytical instruments with efficient molecular screening technology to achieve a method for identifying crude oil and fuel oil that is both fast and accurate and can be widely used on site. This method can not only overcome the shortcomings of existing technologies, but also greatly improve the efficiency of on-site detection and reduce detection costs. Summary of the invention

[0006] The purpose of the present invention is to provide a fast, accurate and easy-to-use oil identification method, which can effectively distinguish crude oil from fuel oil. By combining portable X-ray fluorescence spectrometry analysis technology with microfluidic chip technology, the metal element content and molecular composition in the oil are analyzed to determine the type of oil. The present invention can overcome the problems of complex operation, long analysis time and high cost in the prior art, and has high practical value.

[0007] In order to achieve the above object, the present invention provides a rapid screening method for identifying the commodity attributes of crude oil / fuel oil, comprising the following steps: (1) Take the oil sample to be identified; (2) Use a portable X-ray fluorescence spectrometer to analyze the metal element content in oil samples, especially nickel; (3) Use microfluidic chip technology to analyze the molecular composition of oil samples, especially the molecular chain length and distribution of hydrocarbons in oil products; (4) Based on the analysis results of metal element content and molecular composition, the preset classification model is used to determine whether the oil product is crude oil or fuel oil.

[0008] Furthermore, the step of collecting the oil sample to be identified includes taking samples from oil tanks, pipelines, containers or on-site equipment, and the sample volume is 20-50 mL.

[0009] Furthermore, the analysis of the metal element content is performed using a portable X-ray fluorescence spectrometer, and the measurement time is less than or equal to 10 minutes.

[0010] Furthermore, the metal element is nickel. By establishing a classification standard based on nickel content, when the nickel content is greater than or equal to 50 ppm, the oil product is fuel oil. When the nickel content is less than 50 ppm, it is further judged by molecular composition analysis.

[0011] Furthermore, the molecular composition analysis of the oil sample is carried out using microfluidic chip technology, which includes dynamic screening of hydrocarbon molecules in the sample based on the principle of microcurrent, further determining the oil type by detecting the length and distribution of the molecular chain, and completing data collection within 10 minutes.

[0012] Furthermore, the classification model is a linear regression model, which is used to classify oil types according to the comprehensive analysis results of metal element content and molecular composition.

[0013] Furthermore, the oil type of the oil sample is determined to be "crude oil" or "fuel oil", and a determination value is output through data processing, and the accuracy of the determination value is greater than 95%.

[0014] Furthermore, the rapid screening method is suitable for on-site rapid detection, rapid identification of oil types in industries such as oil fields, refineries, and oil trading, and directly outputs analysis results through mobile terminals or portable devices.

[0015] Furthermore, the establishment of the linear regression model includes the following steps: Establishment of linear regression model The goal of linear regression is to find a best-fitting linear relationship that maps the input features to the classification results of oil products. The process optimizes the regression coefficients by minimizing the sum of squared errors (least squares method).

[0016] The mathematical formula for the linear regression model is: Y=β0+β1X1+β2X2+...+βnXn+ϵ Where Y is the predicted oil type (classification result of crude oil or fuel oil); X1, X2, ..., Xn are input features (such as metal element content, molecular composition characteristics, etc.); β0 is the intercept of the model (constant term); β1, β2, ..., βn are the regression coefficients corresponding to each feature; ϵ is the error term, which represents the difference between the actual result and the predicted result.

[0017] (2) Model training Using the collected sample data, the regression β0, β1, ..., βn is optimized by the least squares method so that the difference between the predicted value of the model and the true value (oil type) is minimized.

[0018] (3) Model verification and evaluation After completing model training, use cross-validation or an independent test set to evaluate the model's performance. Common evaluation metrics include: Accuracy: Determine whether the model prediction results are correct.

[0019] Mean Squared Error (MSE): Evaluates model error. The smaller the error, the better the model.

[0020] Coefficient of determination (R²): Evaluates how well the model fits the data.

[0021] (4) Prediction of oil type Once the regression model is built and its accuracy is verified, it can be used to predict whether a new oil sample is crude oil or fuel oil. By inputting new sample features (such as metal element content and molecular composition features) into the trained regression model, the predicted oil type can be obtained.

[0022] The present invention is beneficial in that: By combining portable X-ray fluorescence spectroscopy technology and microfluidic chip technology, rapid analysis and accurate determination of oil samples are achieved. Compared with the prior art, the present invention not only has the advantages of fast detection speed, simple operation, low cost, etc., but also is suitable for on-site rapid detection and is widely used in oil fields, refineries, oil trade and other industries, and has important practical significance for the rapid identification of oil products. DETAILED DESCRIPTION

[0023] In order to facilitate understanding of the present invention, the present invention is further described below in conjunction with examples. The following examples are only for a better understanding of the present invention and do not mean that the present invention is limited to the following examples.

[0024] Example 1: Rapid identification based on nickel content Sample: Fuel oil nickel content 71.3 ppm; Take 30 mL of the oil sample to be identified with a nickel content of 71.3 ppm, and name the sample as sample 1; (3) Use a portable X-ray fluorescence spectrometer to analyze the metal element content in oil samples.

[0025] Implementation process: Sample 1 was heated to 50°C, shaken thoroughly, and 5 ml was taken for spectral analysis using a portable X-ray fluorescence spectrometer. The content of the metal element nickel in the sample was detected to be 71.1 ppm. The nickel content was greater than 50 ppm, and the sample 1 was determined to be fuel oil.

[0026] Results: Through this method, the sample was accurately judged to be fuel oil in 6 minutes, with an accuracy rate of 100%.

[0027] Example 2: Fuel oil batch detection Experimental conditions: Sample: Fuel oil sample from refinery production line; A 20 mL sample of the fuel oil with a nickel content of 33.7 ppm is named sample 2, and a 20 mL sample of the fuel oil with a nickel content of 85.4 ppm is named sample 3; Use portable X-ray fluorescence spectrometer to analyze the metal element content in oil samples; Use microfluidic chip technology to separate and screen hydrocarbon molecules in samples and detect the length and distribution of molecular chains; Linear regression model was used for classification.

[0028] Implementation process: The samples 2 and 3 were heated to 50°C respectively, and 5 mL was taken after sufficient shaking. The samples were analyzed spectrally using a portable X-ray fluorescence spectrometer. The contents of the metal element nickel in samples 2 and 3 were found to be 33.9 ppm and 85.1 ppm respectively. Sample 3 was judged to be fuel oil based on the nickel content. Since the nickel content of sample 2 was less than 50 ppm, the hydrocarbon molecules in the sample were separated and screened using microfluidic chip technology. The length and distribution of the molecular chains were detected, and the linear regression model was used for classification. Sample 2 was judged to be fuel oil.

[0029] Results: This method can complete the test within 10 minutes, and the oil type of the sample can be accurately determined with an accuracy rate of 98%.

[0030] Example 3: On-site rapid detection of crude oil and fuel oil Experimental conditions: Samples: Crude oil and fuel oil from oil field operations; 20 mL of crude oil sample with a nickel content of 23.6 ppm was sampled and named sample 4, and 20 mL of fuel oil sample with a nickel content of 55.4 ppm was sampled and named sample 5; Use portable X-ray fluorescence spectrometer to analyze the metal element content in oil samples; Use microfluidic chip technology to separate and screen hydrocarbon molecules in samples and detect the length and distribution of molecular chains; Linear regression model was used for classification.

[0031] Implementation process: The samples 4 and 5 were heated to 50°C respectively, and 5 mL was taken after sufficient shaking. The samples were analyzed spectrally using a portable X-ray fluorescence spectrometer. The contents of the metal element nickel in samples 4 and 5 were found to be 23.5 and 55.1 ppm respectively. Sample 5 was judged to be fuel oil based on the nickel content. Since the nickel content of sample 4 was less than 50 ppm, the hydrocarbon molecules in the sample were separated and screened using microfluidic chip technology. The length and distribution of the molecular chains were detected, and the linear regression model was used for classification. Sample 4 was judged to be crude oil.

[0032] Results: The test was completed on site, and the overall analysis time did not exceed 10 minutes. The test results showed that the classification accuracy of the samples was 98%, which is suitable for rapid testing on site in oil fields.

[0033] Example 4: Oil classification based on metal elements and molecular composition Experimental conditions: Samples: Crude oil from different operations; 20 mL of a crude oil sample with a nickel content of 32.1 ppm was sampled and named sample 6, and 20 mL of a crude oil sample with a nickel content of 25.4 ppm was sampled and named sample 7; Use portable X-ray fluorescence spectrometer to analyze the metal element content in oil samples; Use microfluidic chip technology to separate and screen hydrocarbon molecules in samples and detect the length and distribution of molecular chains; Linear regression model was used for classification.

[0034] Implementation process: The samples 6 and 7 were heated to 50°C respectively, and 5 mL was taken out after sufficient shaking. The portable X-ray fluorescence spectrometer was used for spectral analysis. The contents of metal element nickel in samples 6 and 7 were 32.5 and 25.6 ppm respectively. The types of oils of samples 6 and 7 could not be determined based on the nickel content. Microfluidic chip technology was further used to separate and screen the hydrocarbon molecules in the samples, and the length and distribution of the molecular chains were detected. The linear regression model was used for classification. Sample 6 was judged to be crude oil, and sample 7 was judged to be crude oil.

[0035] Results: The proposed method was able to accurately distinguish crude oils from different sources, with a classification accuracy of 97%.

[0036] Example 5: Fuel oil classification based on metal elements and molecular composition Experimental conditions: Samples: fuel oil from different work sites; 30 mL of the fuel oil sample with a nickel content of 43.4 ppm is named sample 8 and 30 mL of the fuel oil sample with a nickel content of 34.6 ppm is named sample 9; Use portable X-ray fluorescence spectrometer to analyze the metal element content in oil samples; Use microfluidic chip technology to separate and screen hydrocarbon molecules in samples and detect the length and distribution of molecular chains; Linear regression model was used for classification.

[0037] Implementation process: The samples 8 and 9 were heated to 50°C respectively, and 5 mL was taken out after sufficient shaking. The portable X-ray fluorescence spectrometer was used for spectral analysis. The contents of metal element nickel in samples 8 and 9 were detected to be 43.9 and 34.1 ppm respectively. The oil types of samples 8 and 9 could not be determined based on the nickel content. The microfluidic chip technology was further used to separate and screen the hydrocarbon molecules in the samples, and the length and distribution of the molecular chains were detected. The linear regression model was used for classification, and it was determined that sample 8 was fuel oil and sample 9 was fuel oil.

[0038] Results: The method was able to accurately distinguish fuel oil from different sources, with a classification accuracy of 96%.

[0039] Example 6: Rapid determination of oil product types in oil trading Experimental conditions: Samples: Crude oil and fuel oil from different traders; Fuel oil: for fuel oil with nickel content of 97.4, 65.3, 44.6, 35.8, and 33.8 ppm, 30 mL samples were taken respectively and named as sample a, sample b, sample c, sample d, and sample e; Crude oil: for crude oil with nickel content of 36.7, 44.6, 21.5, 35.8, and 16.7 ppm, 30 mL samples were taken respectively and named as sample f, sample g, sample h, sample i, and sample j; Use portable X-ray fluorescence spectrometer to analyze the metal element content in oil samples; Use microfluidic chip technology to separate and screen hydrocarbon molecules in samples and detect the length and distribution of molecular chains; Linear regression model was used for classification.

[0040] The samples a, b, c, d, e and f, g, h, i and j were heated to 50°C respectively, 5 mL was taken after sufficient shaking, and spectral analysis was performed using a portable X-ray fluorescence spectrometer. The contents of metal element nickel in the samples were detected to be 97.8, 65.9, 44.7, 35.5, 33.2, 36.3, 44.6, 21.8, 35.2 and 16.1 ppm respectively. Sample a and sample b were judged to be fuel oil based on the nickel content, but the oil types of samples c, d, e, f, g, h, i and j could not be judged based on the nickel content. The hydrocarbon molecules in the samples were further separated and screened by combining the microfluidic chip technology, the length and distribution of the molecular chains were detected, and the linear regression model was used for classification. Sample c, d and e were judged to be fuel oil, and sample f, g, h, i and j were judged to be fuel oil.

[0041] Results: This method can quickly process a large number of samples in oil trade, and the oil type of all samples can be accurately determined with an accuracy rate of 99%. The application of this method in oil trade can effectively improve the efficiency of oil identification.

Claims

1. A rapid screening method for identifying the commodity attributes of crude oil / fuel oil, characterized in that: The following steps are involved: (1) Take the oil sample to be identified; (2) Use a portable X-ray fluorescence spectrometer to analyze the metal element content in oil samples; (3) Use microfluidic chip technology to analyze the molecular composition of oil samples; (4) Based on the analysis results of metal element content and molecular composition, the preset classification model is used to determine whether the oil product is crude oil or fuel oil.

2. A rapid screening method for identifying the properties of crude oil / fuel oil commodities according to claim 1, characterized in that: The step of collecting the oil sample to be identified includes taking samples from oil tanks, pipelines, containers or on-site equipment, and the sample volume is 20-50 mL.

3. A rapid screening method for identifying the commodity attributes of crude oil / fuel oil according to claim 1, characterized in that: The analysis of the metal element content is performed using a portable X-ray fluorescence spectrometer, and the measurement time is less than or equal to 10 minutes.

4. A rapid screening method for identifying the commodity attributes of crude oil / fuel oil according to claim 1, characterized in that: The metal element is nickel, and when the nickel content is greater than or equal to 50 ppm, the oil product is fuel oil, and when the nickel content is less than 50 ppm, it is further determined by molecular composition analysis.

5. A rapid screening method for identifying the commodity attributes of crude oil / fuel oil according to claim 1, characterized in that: The molecular composition analysis of the oil sample is carried out using microfluidic chip technology, which includes dynamic screening of hydrocarbon molecules in the sample based on the principle of microcurrent and completes data collection within 10 minutes.

6. A rapid screening method for identifying the commodity attributes of crude oil / fuel oil according to claim 1, characterized in that: The classification model is a linear regression model.

7. A rapid screening method for identifying the commodity attributes of crude oil / fuel oil according to claim 1, characterized in that: The oil type judgment result of the oil sample is "crude oil" or "fuel oil", and a judgment value is output through data processing, and the accuracy of the judgment value is greater than 95%.