Method, system and storage medium for measuring physical and chemical indicators of diesel based on Raman spectroscopy

Abstract

The present invention discloses a method, system, and storage medium for determining the physical and chemical indicators of diesel fuel based on Raman spectroscopy. The method comprises: iteratively training an original XGBoost model for predicting the physical and chemical indicators of diesel fuel based on the measured physical and chemical indicator values ​​of each diesel fuel sample and a plurality of first Raman spectral data corresponding to each diesel fuel sample to obtain a target XGBoost model; and inputting a plurality of target Raman spectral data of the diesel fuel to be tested into the target XGBoost model to obtain predicted physical and chemical indicator values ​​of the diesel fuel to be tested. Compared with other models for determining the physical and chemical indicators of diesel fuel using Raman spectroscopy, the present invention improves the accuracy of determining the physical and chemical indicators of diesel fuel; and, compared with national standard detection methods, not only can non-destructive determination be performed, but also the speed of determining the physical and chemical indicators of diesel fuel is increased.

Description

Technical Field

[0001] The present invention relates to the technical field of diesel detection, and in particular to a method, system and storage medium for determining physical and chemical indicators of diesel based on Raman spectroscopy. Background Art

[0002] Diesel is a light petroleum product, a complex hydrocarbon mixture, primarily composed of diesel fractions generated through processes such as crude oil distillation, catalytic cracking, thermal cracking, hydrotreating, and petroleum coking. It can also be produced from shale oil processing and coal liquefaction, and is widely used in vehicles and ships. Diesel quality is directly related to its physical and chemical properties, such as cetane number and total sulfur content, which determine its ignition properties, fluidity, and contamination level. Each of these properties has corresponding national standard testing methods (combustion, titration, distillation, filtration, etc.), but the procedures are relatively cumbersome.

[0003] Therefore, it is urgent to provide a technical solution to solve the above technical problems. Summary of the Invention

[0004] To solve the above technical problems, the present invention provides a method, system and storage medium for measuring the physical and chemical indicators of diesel based on Raman spectroscopy.

[0005] The technical solution of the method for determining the physical and chemical indicators of diesel based on Raman spectroscopy of the present invention is as follows:

[0006] Based on the physical and chemical index measurement values ​​of each diesel sample and the multiple first Raman spectrum data corresponding to each diesel sample, the original XGBoost model for diesel physical and chemical index prediction is iteratively trained to obtain a target XGBoost model;

[0007] Multiple target Raman spectral data of the diesel to be tested are input into the target XGBoost model to obtain the predicted values ​​of the physical and chemical indicators of the diesel to be tested.

[0008] The beneficial effects of the method for determining the physical and chemical indicators of diesel based on Raman spectroscopy of the present invention are as follows:

[0009] Compared with other models that use Raman spectroscopy to measure the physical and chemical indicators of diesel, the method of the present invention improves the accuracy of the measurement of the physical and chemical indicators of diesel; at the same time, compared with the national standard detection method, it can not only perform non-destructive measurement, but also improve the speed of the measurement of the physical and chemical indicators of diesel.

[0010] On the basis of the above scheme, the method for determining the physical and chemical indicators of diesel based on Raman spectroscopy of the present invention can be further improved as follows.

[0011] Furthermore, it also includes:

[0012] Acquire multiple raw Raman spectrum data corresponding to each diesel sample;

[0013] Data preprocessing is performed on all original Raman spectrum data corresponding to any diesel sample to obtain a plurality of first Raman spectrum data corresponding to the diesel sample, until a plurality of first Raman spectrum data corresponding to each diesel sample is obtained.

[0014] Furthermore, the step of obtaining a plurality of original Raman spectral data corresponding to each diesel sample includes:

[0015] A handheld spectrometer was used to obtain multiple raw Raman spectral data corresponding to each diesel sample.

[0016] Furthermore, the step of performing data preprocessing on all original Raman spectrum data corresponding to any diesel sample to obtain a plurality of first Raman spectrum data corresponding to the diesel sample includes:

[0017] All original Raman spectral data corresponding to any diesel sample are sequentially subjected to abnormal data cleaning processing, data denoising processing, fluorescence interference subtraction processing, cubic spline interpolation processing and data normalization processing to obtain multiple first Raman spectral data corresponding to the diesel sample.

[0018] Furthermore, the physical and chemical index measurement values ​​include: cetane number measurement value and total sulfur content measurement value; the physical and chemical index prediction values ​​include: cetane number prediction value and total sulfur content prediction value.

[0019] Furthermore, before the step of inputting a plurality of target Raman spectral data of the diesel to be tested into the target XGBoost model to obtain the predicted values ​​of the physical and chemical indicators of the diesel to be tested, the method further includes:

[0020] Acquiring and preprocessing a plurality of original Raman spectrum data of the diesel to be tested to obtain a plurality of target Raman spectrum data of the diesel to be tested;

[0021] The step of inputting a plurality of target Raman spectral data of the diesel to be tested into the target XGBoost model to obtain the predicted values ​​of the physical and chemical indicators of the diesel to be tested comprises:

[0022] Multiple target Raman spectral data of the diesel to be tested are input into the target XGBoost model to predict the physical and chemical indicators of the diesel to be tested, and the predicted value of the cetane number and the predicted value of the total sulfur content of the diesel to be tested are obtained.

[0023] The technical solution of the diesel physical and chemical index determination system based on Raman spectroscopy of the present invention is as follows:

[0024] Includes: training module and measurement module;

[0025] The training module is used to iteratively train the original XGBoost model for predicting the physical and chemical indicators of diesel based on the physical and chemical indicator measurement values ​​of each diesel sample and the multiple first Raman spectrum data corresponding to each diesel sample to obtain a target XGBoost model;

[0026] The determination module is used to input multiple target Raman spectrum data of the diesel to be tested into the target XGBoost model to obtain the predicted values ​​of the physical and chemical indicators of the diesel to be tested.

[0027] The beneficial effects of the diesel physical and chemical index determination system based on Raman spectroscopy of the present invention are as follows:

[0028] Compared with other models that use Raman spectroscopy to measure the physical and chemical indicators of diesel, the system of the present invention improves the accuracy of the measurement of the physical and chemical indicators of diesel; at the same time, compared with the national standard detection method, it can not only perform non-destructive measurement, but also improve the speed of the measurement of the physical and chemical indicators of diesel.

[0029] On the basis of the above scheme, the diesel physical and chemical index determination system based on Raman spectroscopy of the present invention can also be improved as follows.

[0030] Furthermore, it also includes: an acquisition module and a processing module;

[0031] The acquisition module is used to: obtain a plurality of original Raman spectrum data corresponding to each diesel sample;

[0032] The processing module is used to perform data preprocessing on all original Raman spectrum data corresponding to any diesel sample to obtain multiple first Raman spectrum data corresponding to the diesel sample, until multiple first Raman spectrum data corresponding to each diesel sample are obtained.

[0033] Furthermore, the acquisition module is specifically used to:

[0034] A handheld spectrometer was used to obtain multiple raw Raman spectral data corresponding to each diesel sample.

[0035] A technical solution of a storage medium of the present invention is as follows:

[0036] The storage medium stores instructions, and when a computer reads the instructions, the computer is caused to execute the steps of the method for determining the physical and chemical indicators of diesel based on Raman spectroscopy of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 A schematic flow chart of an embodiment of a method for determining physical and chemical indicators of diesel based on Raman spectroscopy provided by the present invention is shown;

[0038] Figure 2The figure shows the validation set effect diagram of the XGBoost model in the embodiment of the method for determining the physical and chemical indicators of diesel based on Raman spectroscopy provided by the present invention;

[0039] Figure 3 A feature ranking diagram of the XGBoost model in an embodiment of the method for determining physical and chemical indicators of diesel based on Raman spectroscopy provided by the present invention is shown;

[0040] Figure 4 A schematic flow chart of steps 101 to 102 of an embodiment of a method for determining physical and chemical indicators of diesel based on Raman spectroscopy provided by the present invention is shown;

[0041] Figure 5 The figure shows a spectrum curve corresponding to a diesel sample in an embodiment of the method for determining physical and chemical indicators of diesel based on Raman spectroscopy provided by the present invention;

[0042] Figure 6 The figure shows a schematic structural diagram of an embodiment of a diesel physical and chemical index determination system based on Raman spectroscopy provided by the present invention. DETAILED DESCRIPTION

[0043] Figure 1 The flowchart of an embodiment of the method for determining the physical and chemical indexes of diesel fuel based on Raman spectroscopy provided by the present invention is shown. Figure 1 As shown, the following steps are included:

[0044] Step 110: Based on the physical and chemical index measurement values ​​of each diesel sample and the plurality of first Raman spectra data corresponding to each diesel sample, iteratively train the original XGBoost model for diesel physical and chemical index prediction to obtain a target XGBoost model;

[0045] Among them, ① the diesel sample is randomly selected diesel, and the values ​​of the physical and chemical indicators corresponding to each diesel sample may be different. ② The physical and chemical indicators include but are not limited to: cetane number and total sulfur content. The cetane number measurement value is obtained according to the national standard GB / T386, and the total sulfur content measurement value is obtained according to the national standard GB / T380. ③ The physical and chemical indicator measurement values ​​include: cetane number measurement value and total sulfur content measurement value. ④ The first Raman spectrum data is: the Raman spectrum data corresponding to the diesel sample, and the first Raman spectrum data is the Raman spectrum data after data preprocessing. ⑤ The original XGBoost model is: the untrained XGBoost model. ⑥ The target XGBoost model is: the trained XGBoost model, which can be used to predict the physical and chemical indicators of diesel.

[0046] It should be noted that ① the XGBoost model selects the gradient boosting tree, the learning rate is 0.03, and the maximum depth of the tree is 5 layers. ② The preset number of iterative training of the XGBoost model defaults to 500 times. ③ The multiple first Raman spectral data of each diesel sample are used as the input X of the XGBoost model, and the physical and chemical index measurement value of each diesel sample is used as Y = [y1, y2]; y1 is the cetane number measurement value, and y2 is the total sulfur content measurement value. ③ The specific training process of the XGBoost model is an existing technology and will not be elaborated here. ④ Among the first Raman spectral data corresponding to all diesel samples, 80% are used for training and 20% are used for verification. The effect of the verification set is as follows Figure 2 As shown. Among them, the mean square error and mean absolute error of the XGB validation cetane number are MSE: 0.3904799473913738 and MAE: 0.18317431858607694 respectively. ⑤ Figure 3 shows the feature ranking graph of XGBoost. Figure 3 In the example, the F-score on the horizontal axis represents the number of times a feature is used, and the vertical axis represents the number of features. The feature importance ranking, which also represents the location of functional groups, directly reflects the substances in the diesel fuel that are relevant to the measured indicator. For example, total sulfur content in diesel fuel does not exist as elemental sulfur, but may exist in groups such as thio, sulfate, and sulfide. Feature ranking can increase the interpretability of the XGBoost model.

[0047] Step 120: Input multiple target Raman spectrum data of the diesel to be tested into the target XGBoost model to obtain predicted values ​​of the physical and chemical indicators of the diesel to be tested.

[0048] Here, ① the diesel to be tested is the diesel for which physical and chemical indicators are to be measured. ② the target Raman spectral data is the Raman spectral data corresponding to the diesel to be tested, which has undergone data preprocessing. ③ the predicted physical and chemical indicators include the predicted cetane number and total sulfur content of the diesel to be tested. ④ the predicted cetane number and total sulfur content of the diesel to be tested are both output by the target XGBoost model.

[0049] It should be noted that the values ​​predicted by the XGBoost model are all predicted values, and the values ​​measured by the national standard method are all measured values ​​or calibrated values; the measured values ​​or calibrated values ​​are used as the Y values ​​during XGBoost model training, and can also be used to calculate the XGBoost model evaluation indicators.

[0050] Preferably, if Figure 4 As shown, it also includes:

[0051] Step 101: Acquire a plurality of original Raman spectrum data corresponding to each diesel sample.

[0052] The original Raman spectral data refers to the Raman spectral data corresponding to the diesel sample without any processing.

[0053] It should be noted that the default number of raw Raman spectral data corresponding to each diesel sample is 200, which can be adjusted according to needs and is not limited here.

[0054] Specifically, step 101 includes:

[0055] A handheld spectrometer was used to obtain multiple raw Raman spectral data corresponding to each diesel sample.

[0056] Among them, ① in this embodiment, the handheld spectrometer uses the Huatai Nuoan 785nm CR2000 handheld spectrometer, and it can also be other models of spectrometers, without limitation. ② By using multiple handheld spectrometers, different laser powers, and different integration times to collect 200 spectra for each diesel sample, 200 original Raman spectrum data corresponding to each diesel sample are obtained. For example, as shown in Table 1 below, the number of diesel samples is 7 (C1-C7), and the number of original Raman spectrum data for each diesel sample is 200. The spectral curve corresponding to each diesel sample is as follows. Figure 5 As shown, the spectral curves from top to bottom are C6, C7, C5, C4, C3, C2, and C1. Figure 5 The horizontal axis represents the Raman shift, and the vertical axis represents the light intensity. Each curve is drawn by averaging all the first Raman spectrum data of the corresponding diesel sample.

[0057] Table 1:

[0058]

[0059] Step 102: performing data preprocessing on all original Raman spectrum data corresponding to any diesel sample to obtain a plurality of first Raman spectrum data corresponding to the diesel sample, until a plurality of first Raman spectrum data corresponding to each diesel sample is obtained.

[0060] Among them, the step of performing data preprocessing on all the original Raman spectrum data corresponding to any diesel sample to obtain multiple first Raman spectrum data corresponding to the diesel sample includes: performing abnormal data cleaning processing, data denoising processing, fluorescence interference subtraction processing, cubic spline interpolation processing and data normalization processing on all the original Raman spectrum data corresponding to the any diesel sample to obtain multiple first Raman spectrum data corresponding to the diesel sample.

[0061] It should be noted that: ① Data denoising was performed using the Savitzky-Golay algorithm. ② Fluorescence interference was subtracted using IAsLS (Improved Asymmetric Least Squares). ③ Data normalization was performed using maximum-minimum normalization. ④ Abnormal data was cleaned using signal-to-noise ratio (SNR) and Pearson coefficient extraction; for example, data with excessive noise or a Pearson coefficient less than 0.75 were discarded.

[0062] Preferably, before step 120, the method further includes:

[0063] A plurality of original Raman spectrum data of the diesel to be tested are acquired and data preprocessed to obtain a plurality of target Raman spectrum data of the diesel to be tested.

[0064] It should be noted that ① the data preprocessing method for the multiple raw Raman spectral data of the diesel to be tested is the same as the data preprocessing method for the multiple raw Raman spectral data of the diesel sample, and will not be elaborated on here. ② The default number of multiple raw Raman spectral data of the diesel to be tested is 50, which can be set according to actual needs and is not limited here.

[0065] Step 120 includes:

[0066] The multiple target Raman spectral data of the diesel to be tested are input into the target XGBoost model to predict the physical and chemical indicators of the diesel, and the predicted value of the cetane number and the measured value of the total sulfur content of the diesel to be tested are obtained.

[0067] It should be noted that the cetane number measurement value and the total sulfur content measurement value of the diesel to be tested are extracted by the national standard testing method, and it is judged that the error between the cetane number measurement value of the diesel to be tested and the cetane number predicted value is less than the preset error threshold, and the error between the total sulfur content measurement value of the diesel to be tested and the total sulfur content measurement value is also less than the preset error threshold, which shows that the XGBoost model used in this embodiment has good accuracy.

[0068] Compared with other models that use Raman spectroscopy to measure the physical and chemical indicators of diesel, the technical solution of this embodiment improves the accuracy of the measurement of the physical and chemical indicators of diesel; at the same time, compared with the national standard detection method, it can not only perform non-destructive measurement, but also improve the speed of the measurement of the physical and chemical indicators of diesel.

[0069] Figure 6 FIG. 1 shows a schematic diagram of an embodiment of a diesel physical and chemical index determination system based on Raman spectroscopy provided by the present invention. Figure 6 As shown, the system 200 includes a training module 210 and a measurement module 220 .

[0070] The training module 210 is used to iteratively train the original XGBoost model for predicting the physical and chemical indicators of diesel based on the physical and chemical indicator measurement values ​​of each diesel sample and the multiple first Raman spectrum data corresponding to each diesel sample to obtain a target XGBoost model;

[0071] The determination module 220 is used to input a plurality of target Raman spectrum data of the diesel to be tested into the target XGBoost model to obtain the predicted values ​​of the physical and chemical indicators of the diesel to be tested.

[0072] Preferably, it further comprises: an acquisition module and a processing module;

[0073] The acquisition module is used to: obtain a plurality of original Raman spectrum data corresponding to each diesel sample;

[0074] The processing module is used to perform data preprocessing on all original Raman spectrum data corresponding to any diesel sample to obtain multiple first Raman spectrum data corresponding to the diesel sample, until multiple first Raman spectrum data corresponding to each diesel sample are obtained.

[0075] The acquisition module is specifically used for:

[0076] A handheld spectrometer was used to obtain multiple raw Raman spectral data corresponding to each diesel sample.

[0077] Compared with other models that use Raman spectroscopy to measure the physical and chemical indicators of diesel, the technical solution of this embodiment improves the accuracy of the measurement of the physical and chemical indicators of diesel; at the same time, compared with the national standard detection method, it can not only perform non-destructive measurement, but also improve the speed of the measurement of the physical and chemical indicators of diesel.

[0078] The above-mentioned parameters and steps for each module to implement corresponding functions in the diesel physical and chemical index measurement system 200 based on Raman spectroscopy in this embodiment can refer to the parameters and steps in the embodiment of the diesel physical and chemical index measurement method based on Raman spectroscopy above, and will not be repeated here.

[0079] An embodiment of the present invention provides a storage medium, comprising: instructions stored in the storage medium, which, when read by a computer, causes the computer to execute the steps of a method for measuring the physical and chemical indicators of diesel based on Raman spectroscopy. For details, reference may be made to the parameters and steps in the embodiment of the method for measuring the physical and chemical indicators of diesel based on Raman spectroscopy described above, which will not be described in detail here.

[0080] Computer storage media such as USB flash drives, mobile hard drives, etc.

[0081] Those skilled in the art will appreciate that the present invention can be implemented as a method, a system, and a storage medium.

[0082] Therefore, the present invention may be embodied in the following forms: entirely in hardware, entirely in software (including firmware, resident software, microcode, etc.), or in a combination of hardware and software, generally referred to herein as a "circuit," "module," or "system." Furthermore, in some embodiments, the present invention may be embodied in the form of a computer program product embodied in one or more computer-readable media, the computer-readable media containing computer-readable program code. Any combination of one or more computer-readable media may be employed. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. Although embodiments of the present invention have been shown and described above, it should be understood that the above embodiments are exemplary and are not intended to limit the present invention. Those skilled in the art may make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for determining the physical and chemical indicators of diesel based on Raman spectroscopy, characterized in that: include: Based on the cetane number and total sulfur content measurements of each diesel sample, as well as multiple first Raman spectra data corresponding to each diesel sample, the original XGBoost model used for predicting diesel physical and chemical indicators is iteratively trained to obtain a target XGBoost model; Multiple target Raman spectral data of the diesel to be tested are input into the target XGBoost model to predict the physical and chemical indicators of the diesel, and obtain the predicted value of the cetane number and the predicted value of the total sulfur content of the diesel to be tested.

2. The method for determining the physical and chemical indicators of diesel based on Raman spectroscopy according to claim 1, wherein: Also includes: Acquire multiple raw Raman spectrum data corresponding to each diesel sample; Data preprocessing is performed on all original Raman spectrum data corresponding to any diesel sample to obtain a plurality of first Raman spectrum data corresponding to the diesel sample, until a plurality of first Raman spectrum data corresponding to each diesel sample is obtained.

3. The method for determining the physical and chemical indicators of diesel based on Raman spectroscopy according to claim 2, characterized in that: The step of obtaining a plurality of original Raman spectral data corresponding to each diesel sample includes: A handheld spectrometer was used to obtain multiple raw Raman spectral data corresponding to each diesel sample.

4. The method for determining the physical and chemical indicators of diesel based on Raman spectroscopy according to claim 2 or 3, characterized in that: The step of performing data preprocessing on all original Raman spectral data corresponding to any diesel sample to obtain a plurality of first Raman spectral data corresponding to the diesel sample includes: All original Raman spectral data corresponding to any diesel sample are sequentially subjected to abnormal data cleaning processing, data denoising processing, fluorescence interference subtraction processing, cubic spline interpolation processing and data normalization processing to obtain multiple first Raman spectral data corresponding to the diesel sample.

5. The method for determining the physical and chemical indicators of diesel based on Raman spectroscopy according to claim 1, wherein: Before the step of inputting a plurality of target Raman spectral data of the diesel to be tested into the target XGBoost model to predict the physical and chemical indicators of the diesel to be tested and obtaining the predicted cetane number and total sulfur content of the diesel to be tested, the method further includes: A plurality of original Raman spectrum data of the diesel to be tested are acquired and data preprocessed to obtain a plurality of target Raman spectrum data of the diesel to be tested.

6. A diesel physical and chemical index determination system based on Raman spectroscopy, characterized in that: include: Training module and measurement module; The training module is used to iteratively train an original XGBoost model for predicting physical and chemical indicators of diesel based on a cetane number measurement value and a total sulfur content measurement value of each diesel sample, as well as a plurality of first Raman spectrum data corresponding to each diesel sample, to obtain a target XGBoost model; The determination module is used to input multiple target Raman spectral data of the diesel to be tested into the target XGBoost model to predict the physical and chemical indicators of the diesel, and obtain the predicted value of the cetane number and the predicted value of the total sulfur content of the diesel to be tested.

7. The diesel physical and chemical index determination system based on Raman spectroscopy according to claim 6, characterized in that: Also includes: Acquisition module and processing module; The acquisition module is used to: obtain a plurality of original Raman spectrum data corresponding to each diesel sample; The processing module is used to perform data preprocessing on all original Raman spectrum data corresponding to any diesel sample to obtain multiple first Raman spectrum data corresponding to the diesel sample, until multiple first Raman spectrum data corresponding to each diesel sample are obtained.

8. The diesel physical and chemical index determination system based on Raman spectroscopy according to claim 7, characterized in that: The acquisition module is specifically used for: A handheld spectrometer was used to obtain multiple raw Raman spectral data corresponding to each diesel sample.

9. A storage medium, characterized in that: The storage medium stores instructions, and when a computer reads the instructions, the computer is caused to execute the method for determining physical and chemical indicators of diesel based on Raman spectroscopy according to any one of claims 1 to 5.

Images