Near-infrared spectrum interference correction method and system for vehicle gasoline detection

Abstract

The present invention relates to the technical field of gasoline detection, and in particular to a near-infrared spectrum interference correction method and system for detecting automotive gasoline. When correcting the near-infrared spectrum of automotive gasoline, the present invention collects near-infrared spectrum data of interference components in gasoline, analyzes characteristic bands of the interference components, then determines whether the characteristic bands of target components and interference components overlap to determine candidate interference bands, and determines different interference correction methods based on the interference indexes of the corresponding candidate interference bands, that is, adopts different interference correction methods for different candidate interference correction bands, thereby improving the pertinence and accuracy of the interference correction.

Description

Technical Field

[0001] The present invention relates to the technical field of gasoline detection, and in particular to a near-infrared spectrum interference correction method and system for detecting gasoline for vehicles. Background Art

[0002] As an important energy product, the quality inspection of motor gasoline is of great significance for ensuring engine performance and environmental protection. Methanol and ethanol are common gasoline additives, and the accurate determination of their content is crucial for evaluating gasoline quality and safety. Near-infrared spectroscopy is a rapid, non-destructive and efficient analytical technique widely used in the field of chemical analysis. However, in the near-infrared spectral analysis of motor gasoline, the spectral signals of methanol and ethanol may be interfered with by other components, resulting in a decrease in the accuracy of the quantitative analysis of methanol and ethanol. Therefore, the development of an effective near-infrared spectral interference correction method is of great significance for improving the accuracy of the determination of methanol and ethanol content in motor gasoline.

[0003] There are schemes for interference correction of near-infrared spectra in the prior art. For example, a Chinese invention patent application (CN117191739A) discloses a method and system for near-infrared spectroscopy quantitative detection that corrects for moisture interference. The method includes: constructing a training set of samples, selecting a moisture correction set and a test set from the training set; using the moisture correction set to calculate a moisture correction factor using an external parameter orthogonalization algorithm; collecting the near-infrared spectrum of the training set, correcting its near-infrared spectrum using the moisture correction factor to obtain a moisture correction spectrum, extracting the moisture correction spectrum using a wavelength selection algorithm to obtain a characteristic spectrum, regressing the characteristic spectrum and the chemical component values to be analyzed to establish a moisture correction quantitative detection model; extracting the characteristic spectrum of the sample in the test set, inputting it into the moisture correction quantitative detection model, and predicting the component content; however, the above scheme is only a conventional correction method using a model, resulting in low correction accuracy. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention provides a near-infrared spectral interference correction method and system for automotive gasoline detection, which are used to solve the problems existing in the prior art.

[0005] According to one aspect of the present invention, a near-infrared spectrum interference correction method for detecting vehicle gasoline is provided, which is used to perform interference correction on the near-infrared spectrum for detecting the methanol and ethanol content of vehicle gasoline, comprising the following steps:

[0006] S1: Obtain motor gasoline sample;

[0007] S2: collecting spectral data of the gasoline sample to obtain original spectral data of the gasoline;

[0008] S3: Obtain interference information of interfering components in motor gasoline;

[0009] S4: Determine the characteristic bands of the near-infrared spectra of methanol and ethanol;

[0010] S5: performing a spectral overlap operation on the characteristic band of the near-infrared spectrum of methanol, the characteristic band of the near-infrared spectrum of ethanol, and the characteristic band of the interfering component to obtain a candidate interfering band;

[0011] S6: Evaluate the interference level of the candidate interference band to obtain an interference level evaluation result;

[0012] S7: applying different near-infrared spectrum preprocessing methods to the spectral data of the candidate interference bands of the original spectral data of the motor gasoline according to the interference degree evaluation result to achieve interference correction.

[0013] Preferably, in S2, a NIR-501A portable Fourier near-infrared oil analyzer is used to collect near-infrared spectra, with a spectral range of 9100 to 4000 cm -1 , wave number accuracy: 6523±1 cm -1 , wave number accuracy: ±0.1cm -1 The signal-to-noise ratio is better than 10000:1. The data are collected at three different directions, and the average spectrum is taken as the original spectrum data of the automotive gasoline sample.

[0014] Preferably, the S3 is specifically:

[0015] S3.1: Collect near-infrared spectral data of interfering components;

[0016] S3.2: Determine a characteristic band of the near-infrared spectrum data of the interference component as interference information.

[0017] Preferably, in S6, the degree of interference is quantified by calculating the overlapping area and the absorption peak intensity ratio of the characteristic band of the target component and the characteristic band of the potential interfering component, wherein the interference index is defined as an indicator for quantifying the degree of interference, and the calculation formula of the interference index II is:

[0018] ;

[0019] Where, A target ( λ ) and A interferent ( λ ) are the target component and the interfering component at wavelengths λ The absorption intensity at , overlap represents the overlapping band range of the target component and the interfering component, and targetband represents the characteristic band range of the target component.

[0020] Preferably, the S7 is specifically:

[0021] If the interference level evaluation result is less than 0.2, a first spectrum correction method is adopted for the candidate interference band; otherwise, a second spectrum correction method is adopted for the candidate interference band.

[0022] Preferably, the first spectral correction method is: performing baseline correction and smoothing processing on the corresponding candidate interference bands in the original spectral data of the motor gasoline; the second spectral correction method is: performing baseline correction, smoothing processing and correction using an interference correction model on the corresponding candidate interference bands in the original spectral data of the motor gasoline.

[0023] Preferably, the interference correction model is a hybrid model of a convolutional neural network model and a long short-term memory network model (LSTM).

[0024] Preferably, the correction process using the interference correction model is specifically as follows:

[0025] Sa: Construct interference correction model;

[0026] Sb: Design the loss function of the interference correction model;

[0027] Sc: Inputting the candidate correction wavebands of the original spectral data of the motor gasoline and the interference information into the interference correction model to obtain a correction result.

[0028] Preferably, the loss function is a composite loss function of root mean square error and cosine similarity impairment,

[0029] The formula for the mean square error is:

[0030] ;

[0031] Where N is the number of data sets, y i is the true value of the i-th sample, is the predicted value of the i-th sample;

[0032] The formula for cosine similarity damage is:

[0033] ;

[0034] Among them, the formula of the loss function CompositeLoss is:

[0035] ;

[0036] Where α and β are weight parameters.

[0037] According to another aspect of the present invention, a near-infrared spectral interference correction system for vehicle gasoline detection is provided. The system adopts the above-mentioned near-infrared spectral interference correction method for vehicle gasoline detection. The system comprises:

[0038] A sample acquisition module, used for acquiring motor gasoline samples;

[0039] A raw data acquisition module, configured to acquire spectral data of the gasoline sample to obtain raw spectral data of the gasoline;

[0040] Interference information acquisition module, used to obtain interference information of interfering components in motor gasoline;

[0041] A characteristic band determination module is used to determine the characteristic bands of the near-infrared spectra of methanol and ethanol;

[0042] a candidate interference band determination module, configured to perform a spectral overlap operation on the characteristic bands of the near-infrared spectrum of methanol, the characteristic bands of the near-infrared spectrum of ethanol, and the characteristic bands of the interference components to obtain candidate interference bands;

[0043] An interference level assessment module is used to assess the interference level of the candidate interference bands and obtain an interference level assessment result;

[0044] The interference correction module is used to adopt different near-infrared spectrum preprocessing methods to the spectral data of the candidate interference bands of the original spectral data of the motor gasoline according to the interference degree evaluation result, so as to achieve interference correction.

[0045] The present invention has the following technical effects:

[0046] When correcting the near-infrared spectrum of automotive gasoline, the present invention collects near-infrared spectrum data of interference components in gasoline, analyzes characteristic bands of the interference components, then determines whether the characteristic bands of the target component and the interference component overlap to determine candidate interference bands, and determines different interference correction methods based on the interference index of the corresponding candidate interference bands, that is, adopts different interference correction methods for different candidate interference correction bands, thereby improving the pertinence and accuracy of interference correction.

[0047] Furthermore, this embodiment uses a deep learning model to correct interference, introducing interference information as an auxiliary input during the model input process, enabling the model to identify and correct interference signals. For example, in a CNN-LSTM hybrid model, the spectral characteristics of the interfering component can be input into the model along with the raw spectral data of motor gasoline. The model automatically identifies and corrects the interference signal by learning the characteristic patterns in the data, thereby achieving interference correction and improving the accuracy of interference correction. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0049] Figure 1 This is a flow chart of a near-infrared spectroscopy interference correction method for detecting automotive gasoline provided by an embodiment of the present invention;

[0050] Figure 2 This is a flowchart of correction using an interference correction model provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0051] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention are described clearly and completely below. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are also within the scope of protection of the present invention.

[0052] Example 1, attached Figure 1 A flow chart of a near infrared spectroscopy interference correction method for detecting gasoline for automobiles is shown in the attached figure. Figure 1 As shown, the near-infrared spectrum interference correction method for detecting automotive gasoline is used to perform interference correction on the near-infrared spectrum for detecting methanol and ethanol content in automotive gasoline, comprising the following steps:

[0053] S1: Obtain motor gasoline sample;

[0054] Ten different batches of automotive gasoline samples were collected from a gas station. The samples were filtered using a 0.45 μm filter membrane to remove impurities, thereby obtaining automotive gasoline samples.

[0055] S2: collecting spectral data of the gasoline sample to obtain original spectral data of the gasoline;

[0056] The NIR-501A portable Fourier near-infrared oil analyzer was used to collect near-infrared spectra with a spectral range of 9100–4000 cm -1 , wave number accuracy: 6523±1 cm -1 , wave number accuracy: ±0.1cm -1 The signal-to-noise ratio is better than 10000:1. The data are collected at three different directions, and the average spectrum is taken as the original spectrum data of the automotive gasoline sample.

[0057] S3: Obtain interference information of interfering components in motor gasoline;

[0058] In fact, the interfering components in motor gasoline mainly include aromatic hydrocarbons and olefins, which may interfere with the detection of methanol and ethanol content. Therefore, this embodiment mainly achieves the purpose of motor gasoline near-infrared spectrum correction by removing the spectral interference information of aromatic hydrocarbons and olefins.

[0059] Wherein, the S3 is specifically:

[0060] S3.1: Collect near-infrared spectral data of interfering components;

[0061] The interfering components in motor gasoline mainly include aromatic hydrocarbons and olefins. Therefore, in this example, benzene is used as the aromatic interfering component, and ethylene is used as the olefin interfering component. Pure benzene samples and pure ethylene samples are prepared for near-infrared spectral data acquisition respectively.

[0062] Similarly, the NIR-501A portable Fourier near-infrared oil analyzer was used to collect near-infrared spectra with a spectral range of 9100–4000 cm -1 , wave number accuracy: 6523±1 cm -1 , wave number accuracy: ±0.1cm -1 The signal-to-noise ratio is better than 10000:1. The data are collected at three different directions and the average spectrum is taken as the near-infrared spectral data of the interference component.

[0063] S3.2: Determine a characteristic band of the near-infrared spectrum data of the interference component as interference information;

[0064] In this step, the characteristic bands of the near-infrared spectral data of the pure aromatic hydrocarbons and pure olefins are determined by reading literature; and the characteristic bands of the near-infrared spectral data of the pure aromatic hydrocarbons and pure olefins are used as interference information.

[0065] The characteristic bands of the aromatic hydrocarbons (such as benzene) in the near-infrared spectrum appear at 1400-1500 nm and 1600-1700 nm, and the characteristic bands of the olefins (such as ethylene) appear at 1600-1700 nm and 2200-2300 nm.

[0066] S4: Determine the characteristic bands of the near-infrared spectra of methanol and ethanol;

[0067] In this step, near-infrared spectra of pure methanol and pure ethanol are collected respectively, and the collection parameters are the same as those in the above step, thereby obtaining near-infrared spectral data of pure methanol and pure ethanol, and then the characteristic bands of the near-infrared spectra of methanol and ethanol are determined by searching the literature.

[0068] S5: performing a spectral overlap operation on the characteristic band of the near-infrared spectrum of methanol, the characteristic band of the near-infrared spectrum of ethanol, and the characteristic band of the interfering component to obtain a candidate interfering band;

[0069] By overlaying the spectra, we observe the overlap between the characteristic bands of the target components (methanol and ethanol) and the characteristic bands of the interfering components. For example, if the characteristic absorption peaks of aromatic hydrocarbons overlap with those of methanol near 1400-1500 nm, there may be interference in this band. Therefore, the 1400-1500 nm characteristic band is selected as a candidate interfering band. It is worth emphasizing that the candidate correction bands can be one or more.

[0070] S6: Evaluate the interference level of the candidate interference band to obtain an interference level evaluation result;

[0071] In this step, the interference degree is quantified by calculating the overlapping area of the characteristic band of the target component and the characteristic band of the potential interfering component, the absorption peak intensity ratio and other indicators. The Interference Index (II) is defined as an indicator for quantifying the interference degree.

[0072] The calculation formula of interference index II is:

[0073] ;

[0074] Where, A target ( λ )and A interferent ( λ ) are the target component and the interfering component at wavelengths λ The absorption intensity at , overlap represents the overlapping band range of the target component and the interfering component, and targetband represents the characteristic band range of the target component.

[0075] S7: applying different near-infrared spectrum preprocessing methods to the spectral data of the candidate correction bands of the original spectral data of the motor gasoline according to the interference degree evaluation result, so as to achieve interference correction.

[0076] Wherein, the S7 is specifically:

[0077] If the interference index of the candidate correction band is less than 0.2, the first spectral correction method is applied to the candidate correction band; otherwise, the second spectral correction method is applied to the candidate correction band;

[0078] The first spectrum correction method is: performing baseline correction and smoothing on the corresponding candidate correction bands in the original spectrum data of the motor gasoline;

[0079] The baseline correction is to perform baseline correction on the spectral data using the moving average method to remove the background signal, and the smoothing is to perform smoothing on the spectral data using the moving average method or the Savitzky-Golay smoothing method to reduce the influence of noise;

[0080] The second spectrum correction method comprises: performing baseline correction, smoothing processing and correction using an interference correction model on the corresponding candidate correction bands in the original spectrum data of the motor gasoline;

[0081] In this step, this embodiment proposes a hybrid model (CNN-LSTM) of a convolutional neural network model (CNN) and a long short-term memory network model (LSTM) to construct an interference correction model. The convolutional neural network model can automatically extract local features in the spectral data, while the long short-term memory network model can capture the time series features and interactions between components in the spectral data. Through the combination of CNN and LSTM, the model can more comprehensively handle the complexity of the spectral data.

[0082] As attached Figure 2 As shown, the correction process using the interference correction model is specifically as follows:

[0083] Sa: Construct interference correction model;

[0084] Among them, the convolutional neural network model part adopts a multi-layer convolution layer and pooling layer structure to extract the local features of spectral data; the long short-term memory network model part adopts a multi-layer LSTM unit to capture the time series characteristics of spectral data, and feature fusion is performed between the convolutional neural network model and the long short-term memory network model through a fully connected layer.

[0085] Sb: Design the loss function of the interference correction model;

[0086] To improve the model's calibration accuracy, this example proposes a composite loss function that combines mean squared error (MSE) and cosine similarity loss. MSE measures the difference between the model output and the target value, while cosine similarity loss measures the directional consistency between the model output and the target value. This composite loss function allows the model to simultaneously optimize both the amplitude and shape characteristics of spectral data.

[0087] The formula for the mean square error is:

[0088] ;

[0089] Where N is the number of data sets, y i is the true value of the i-th sample, is the predicted value of the i-th sample;

[0090] The formula for cosine similarity damage is:

[0091] ;

[0092] Among them, the formula of the loss function CompositeLoss is:

[0093] ;

[0094] Where α and β are weight parameters used to balance the importance of the two loss functions.

[0095] Sc: Inputting the corresponding candidate correction bands in the original spectral data of motor gasoline and the interference information into the interference correction model to obtain a correction result.

[0096] In this step, interference information is introduced as an auxiliary input during the model input process, enabling the model to identify and correct interference signals. For example, in a CNN-LSTM hybrid model, the spectral characteristics of the interfering component can be input into the model along with the original spectral data of motor gasoline. The model automatically identifies and corrects the interference signal by learning the characteristic patterns in the data, thereby achieving interference correction and improving the accuracy of interference correction. Compared with the traditional method of inputting only the spectral data to be corrected into the model, the model input method of this embodiment reduced the RMSE on the test set by approximately 30%, reduced the MAE by approximately 25%, and improved the R² by approximately 5%. This shows that the model input method of this embodiment has significant advantages in the prediction accuracy and reliability of spectral data.

[0097] In Example 2, the present invention further provides a near-infrared spectrum interference correction system for detecting gasoline for vehicles. The system adopts a near-infrared spectrum interference correction method for detecting gasoline for vehicles in Example 1. The system includes:

[0098] A sample acquisition module, used for acquiring motor gasoline samples;

[0099] A raw data acquisition module, configured to acquire spectral data of the gasoline sample to obtain raw spectral data of the gasoline;

[0100] Interference information acquisition module, used to obtain interference information of interfering components in motor gasoline;

[0101] A characteristic band determination module is used to determine the characteristic bands of the near-infrared spectra of methanol and ethanol;

[0102] a candidate interference band determination module, configured to perform a spectral overlap operation on the characteristic bands of the near-infrared spectrum of methanol, the characteristic bands of the near-infrared spectrum of ethanol, and the characteristic bands of the interference components to obtain candidate interference bands;

[0103] An interference level assessment module is used to assess the interference level of the candidate interference bands and obtain an interference level assessment result;

[0104] The interference correction module is used to adopt different near-infrared spectrum preprocessing methods to the spectral data of the candidate correction bands of the original spectral data of the motor gasoline according to the interference degree evaluation result, so as to achieve interference correction.

[0105] Example 3: The present invention also provides an electronic device, including one or more processors and a memory.

[0106] The processor may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.

[0107] The memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), a hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor may execute the program instructions to implement the near-infrared spectral interference correction method for automotive gasoline detection described in any embodiment of the present application as described above and / or other desired functions. Various contents such as initial external parameters and threshold values may also be stored in the computer-readable storage medium.

[0108] In one example, the electronic device may further include an input device and an output device, these components interconnected via a bus device and / or other connection mechanism (not shown). The input device may include, for example, a keyboard, a mouse, etc. The output device may output various information to the outside, including warning information, braking force, etc. The output device may include, for example, a display, a speaker, a printer, a communication network, and remote output devices connected thereto.

[0109] Of course, for the sake of simplicity, components such as buses, input / output interfaces, etc. are omitted. In addition, the electronic device may further include any other appropriate components according to specific application scenarios.

[0110] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to implement the functions of a near-infrared spectral interference correction method for automotive gasoline detection provided by any embodiment of the present application.

[0111] The computer program product may be written in any combination of one or more programming languages to implement the program code for performing the operations of the embodiments of the present application, including object-oriented programming languages such as Java, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0112] In addition, an embodiment of the present application may also be a computer-readable storage medium having computer program instructions stored thereon. When the computer program instructions are executed by a processor, the processor implements a near-infrared spectral interference correction method for vehicle gasoline detection provided in any embodiment of the present application.

[0113] The computer-readable storage medium may be any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor device, apparatus, or component, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0114] It should be noted that the terms used in the present invention are only for describing specific embodiments and are not intended to limit the scope of this application. As shown in the present specification, unless the context clearly indicates an exception, the words "one", "a", "a kind of" and / or "the" do not specifically refer to the singular and may also include the plural. The terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method or device comprising a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method or device. In the absence of further restrictions, the elements defined by the sentence "comprise a..." do not exclude the presence of other identical elements in the process, method or device comprising the elements.

[0115] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A near-infrared spectrum interference correction method for detecting gasoline for automobiles, which is used to perform interference correction on the near-infrared spectrum for detecting methanol and ethanol content in gasoline for automobiles, characterized in that: include: S1: Obtain motor gasoline sample; S2: collecting spectral data of the gasoline sample to obtain original spectral data of the gasoline; S3: Obtain interference information of interfering components in motor gasoline; S4: Determine the characteristic bands of the near-infrared spectra of methanol and ethanol; S5: performing a spectral overlap operation on the characteristic band of the near-infrared spectrum of methanol, the characteristic band of the near-infrared spectrum of ethanol, and the characteristic band of the interfering component to obtain a candidate interfering band; S6: Evaluate the interference level of the candidate interference band to obtain an interference level evaluation result; S7: applying different near-infrared spectrum preprocessing methods to the spectral data of the candidate interference bands of the original spectral data of the motor gasoline according to the interference degree assessment result to achieve interference correction; S7 specifically comprises: If the interference degree evaluation result is less than 0.2, the first spectrum correction method is applied to the candidate interference band; otherwise, the second spectrum correction method is applied to the candidate interference band; The first spectral correction method is: performing baseline correction and smoothing processing on the corresponding candidate interference bands in the original spectral data of the motor gasoline; the second spectral correction method is: performing baseline correction, smoothing processing and correction using an interference correction model on the corresponding candidate interference bands in the original spectral data of the motor gasoline.

2. The near-infrared spectroscopy interference correction method for motor gasoline detection according to claim 1, characterized in that: In S2, a NIR-501A portable Fourier near-infrared oil analyzer was used to collect near-infrared spectra with a spectral range of 9100 to 4000 cm -1 , wave number accuracy: 6523±1 cm -1 , wave number accuracy: ±0.1cm -1 The signal-to-noise ratio is better than 10000:

1. The data are collected at three different directions, and the average spectrum is taken as the original spectrum data of the automotive gasoline sample.

3. The near-infrared spectrum interference correction method for vehicle gasoline detection according to claim 1, characterized in that: The S3 is specifically: S3.1: Collect near-infrared spectral data of interfering components; S3.2: Determine a characteristic band of the near-infrared spectrum data of the interference component as interference information.

4. The near-infrared spectroscopy interference correction method for vehicle gasoline detection according to claim 1, characterized in that: In S6, the degree of interference is quantified by calculating the overlapping area and the absorption peak intensity ratio of the characteristic band of the target component and the characteristic band of the potential interfering component, wherein the interference index is defined as an indicator for quantifying the degree of interference, and the calculation formula of the interference index II is: ； Where, A target ( λ )and A interferent ( λ ) are the target component and the interfering component at wavelengths λ The absorption intensity at , overlap represents the overlapping band range of the target component and the interfering component, and targetband represents the characteristic band range of the target component.

5. The near-infrared spectrum interference correction method for vehicle gasoline detection according to claim 1, characterized in that: The interference correction model is a hybrid model of a convolutional neural network model and a long short-term memory network model.

6. The near-infrared spectrum interference correction method for vehicle gasoline detection according to claim 5, characterized in that: The specific process of using the interference correction model for correction is as follows: Sa: Construct interference correction model; Sb: Design the loss function of the interference correction model; Sc: Inputting the candidate correction wavebands of the original spectral data of the motor gasoline and the interference information into the interference correction model to obtain a correction result.

7. The near-infrared spectrum interference correction method for vehicle gasoline detection according to claim 6, characterized in that: The loss function is a composite loss function of root mean square error and cosine similarity damage, Among them, the formula for mean square error MSE is: ； Where N is the number of data sets, y i is the true value of the i-th sample, is the predicted value of the i-th sample; Among them, the formula of cosine similarity loss is: ； Among them, the formula of the composite loss function CompositeLoss is: ； Where α and β are weight parameters.

8. A near-infrared spectroscopy interference correction system for automotive gasoline detection, characterized in that: The system adopts a near-infrared spectrum interference correction method for vehicle gasoline detection according to any one of claims 1 to 7, and the system comprises: A sample acquisition module, used for acquiring motor gasoline samples; A raw data acquisition module, configured to acquire spectral data of the gasoline sample to obtain raw spectral data of the gasoline; Interference information acquisition module, used to obtain interference information of interfering components in motor gasoline; A characteristic band determination module is used to determine the characteristic bands of the near-infrared spectra of methanol and ethanol; a candidate interference band determination module, configured to perform a spectral overlap operation on the characteristic bands of the near-infrared spectrum of methanol, the characteristic bands of the near-infrared spectrum of ethanol, and the characteristic bands of the interference components to obtain candidate interference bands; An interference level assessment module is used to assess the interference level of the candidate interference bands and obtain an interference level assessment result; The interference correction module is used to adopt different near-infrared spectrum preprocessing methods to the spectral data of the candidate interference bands of the original spectral data of the motor gasoline according to the interference degree evaluation result, so as to achieve interference correction.

Images