Sea surface oil film thickness detection method based on laser fluorescence spectrum and multiband fluorescence index

Abstract

The invention relates to a sea surface oil film thickness detection method based on a laser fluorescence spectrum and a multiband fluorescence index, which comprises the following steps: acquiring fluorescence spectrum data of an oil film of a sample to be detected, and taking an average value of every N spectra as an original fluorescence spectrum of the oil film sample; carrying out pretreatment on the original fluorescence spectrum summer falsification; for a to-be-detected sample, calculating fluorescence spectrum indexes of an oil film of the detected sample based on different types of fluorescence index calculation formulas; based on the fluorescence spectrum index of the oil film of the detection sample, constructing different types of oil film thickness prediction models; and based on the evaluation indexes, determining different types of oil film thickness prediction models and corresponding optimal emission wave band combinations, thereby realizing oil film thickness detection. An oil film thickness prediction model is established by utilizing a quantitative analysis method and combining oil spill simulation experiment data, and fluorescence spectrum analysis is used as a non-contact and high-sensitivity detection means, so that the thickness change of the oil film can be reflected through the fluorescence characteristic of the oil film.

Description

Technical Field

[0001] The invention belongs to the technical field of sea surface oil spill detection and relates to a sea surface oil film thickness detection method based on laser fluorescence spectrum and multi-band fluorescence index. Background Art

[0002] With the rapid development of global oil resources and the shipping industry, the frequency of marine oil spills is increasing. After an oil spill, it spreads across the sea surface, forming an oil film that affects phytoplankton photosynthesis and reduces oxygen levels in seawater. Toxic substances in oil can cause the death of marine life or reduce reproductive capacity. This not only severely damages the marine ecosystem but also has a significant negative impact on the economic and social development of coastal areas. Rapid and accurate detection and identification of oil spills is crucial for oil spill response and protecting the marine ecosystem. In oil spills, oil film thickness is a key indicator for assessing the extent of pollution and ecological damage. However, traditional oil film thickness measurement methods have numerous limitations. For example, laser triangulation requires complex optical equipment and is susceptible to marine environmental influences; microwave radiation methods have low resolution for oil film thickness; and laser ultrasonic methods require complex equipment and are costly. In practical applications, these methods are often limited by the physical and chemical properties of the oil film, instrument performance, and environmental conditions, resulting in inaccurate measurement results. Summary of the Invention In order to solve the above problems, the technical solution adopted by the present invention is: a method for detecting the thickness of sea surface oil film based on laser fluorescence spectrum and multi-band fluorescence index, comprising the following steps: Obtain fluorescence spectrum data of the oil film sample to be tested, and take the average value of every N spectra as the original fluorescence spectrum of the oil film sample; Preprocess the raw fluorescence spectrum data; For the samples to be tested, the fluorescence spectrum index of the sample oil film is calculated based on different types of fluorescence index calculation formulas; Based on the fluorescence spectrum index of the oil film of the detected sample, different types of oil film thickness prediction models are constructed; Based on the evaluation index, the optimal emission band combination corresponding to different types of oil film thickness prediction models is determined, thereby realizing the detection of oil film thickness.

[0003] Furthermore: the calculation formulas for the different types of fluorescence indices are as follows:

[0004]

[0005]

[0006]

[0007] in: , , Indicates the fluorescence intensity corresponding to emission wavelengths a, b and c in the 370-700nm spectral range, , EX is the excitation wavelength of 355 nm.

[0008] Furthermore, the process of constructing different types of oil film thickness prediction models based on the fluorescence spectrum index of the sample oil film is as follows: S41: For each type of TBFI fluorescence index, all bands within the fluorescence spectrum are traversed to calculate the TBFI value under each band combination, and the TBFI value under each band combination is subjected to partial least squares regression with the oil film thickness, and finally the determination coefficient under each band combination is calculated; S42: Arrange the determination coefficients of all band combinations in descending order, and select the band combination with the largest determination coefficient; S43: Combine the band combination with the largest determination coefficient with partial least squares regression to establish a TBFI-PLSR model, determine the independent variable group (TBFI) and the dependent variable group (oil film thickness), extract the first principal component of the two variable groups, establish the regression of the two variable groups on the principal component, replace the residual matrix, repeat the above steps to obtain the regression equation, perform cross-validation, and predict the oil film thickness.

[0009] Furthermore: the evaluation indicators include determination coefficient, root mean square error and mean square relative error.

[0010] Furthermore: the calculation formula of the determination coefficient is as follows:

[0011] Where n is the number of samples, is the actual observed value of the i-th sample, i.e., the true value. The model prediction value of the i-th sample, is the average of all actual observations.

[0012] A device for detecting sea surface oil film thickness based on laser fluorescence spectrum and multi-band fluorescence index includes: Laser: As an excitation light source, used to illuminate the sample to stimulate fluorescence signals; Power supply: used to power the laser; Spectrometer: used to collect fluorescence spectrum signals; PC: used to display the collected fluorescence spectrum signal; The PC is connected to the spectrometer via an RJ-45 interface; Detection module: used for receiving the spectrum signal collected by the spectrometer and detecting the thickness of the sea surface oil film based on the method described.

[0013] The present invention provides a method for detecting sea surface oil film thickness based on laser fluorescence spectroscopy and multi-band fluorescence index. This rapid, accurate, and adaptable method is highly suitable for complex environments and holds great significance. Based on sea surface oil spill experiments, fluorescence spectral data corresponding to different oil film thicknesses is obtained. A quantitative analysis method, combined with the data from oil spill simulation experiments, is used to establish an oil film thickness prediction model. Fluorescence spectroscopy, a non-contact, highly sensitive detection method, can reflect changes in oil film thickness through the fluorescence properties of the oil film. Based on this technical principle, the present invention proposes a method for detecting sea surface oil film thickness based on fluorescence index. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the 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 labor.

[0015] Figure 1 is a schematic diagram of the experimental device of the present invention; Figure 2 Flow chart for the implementation of the present invention; Figure 3 This is a flow chart of the optimal band combination algorithm of the present invention; Figure 4 The prediction results of 92# gasoline are shown in Figure 1, where (a) is a scatter plot of the measured and predicted values ​​of gasoline oil film thickness based on the TBFI1-PLSR model, (b) is a scatter plot of the measured and predicted values ​​of gasoline oil film thickness based on the TBFI2-PLSR model, (c) is a scatter plot of the measured and predicted values ​​of gasoline oil film thickness based on the TBFI3-PLSR model, and (d) is a scatter plot of the measured and predicted values ​​of gasoline oil film thickness based on the TBFI4-PLSR model. DETAILED DESCRIPTION It should be noted that, unless there is any conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0016] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is in no way intended to limit the present invention and its application or use. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0017] Figure 1 is a schematic diagram of the experimental device of the present invention; A device for detecting sea surface oil film thickness based on laser fluorescence spectrum and multi-band fluorescence index: comprising: Laser: As an excitation light source, it is used to illuminate the sample to excite the fluorescence signal. The excitation wavelength of the laser is 355nm; Power supply: used to power the laser; Spectrometer: used to collect fluorescence spectrum signal data; a total of 330 bands are collected; PC: used to display the collected fluorescence spectrum signal; the PC is connected to the spectrometer via an RJ-45 interface; Detection module: used for receiving the spectrum signal collected by the spectrometer and detecting the thickness of the sea surface oil film based on the method described.

[0018] Figure 2 Flow chart for the implementation of the present invention; A method for detecting sea surface oil film thickness based on laser fluorescence spectroscopy and multi-band fluorescence index includes the following steps: S1: Obtain the fluorescence spectrum data of the oil film sample to be tested, and take the average value of every N spectra as the original fluorescence spectrum of the oil film sample; N can be 10; S2: Preprocess the original fluorescence spectrum; S3: For the sample to be tested, the fluorescence spectrum index of the oil film of the sample is calculated based on different types of fluorescence index calculation formulas; S4: Based on the fluorescence spectrum index of the oil film of the detected sample, different types of oil film thickness prediction models are constructed; S5: Based on the evaluation index, the optimal emission band combination corresponding to different types of oil film thickness prediction models is determined, thereby realizing the detection of oil film thickness.

[0019] Steps S1 / S2 / S3 / S4 / S5 are performed sequentially; The raw fluorescence spectrum data was preprocessed, including smoothing and filtering, to remove noise interference. The important characteristic spectral bands were selected, that is, the characteristic spectral bands within the range of 370-700nm were selected as the fluorescence spectrum data for subsequent analysis.

[0020] The calculation formulas for the different types of fluorescence indices are as follows:

[0021]

[0022]

[0023]

[0024] in: , , Indicates the fluorescence intensity corresponding to emission wavelengths a, b and c in the 370-700nm spectral range, , EX is the excitation wavelength of 355 nm.

[0025] Figure 3 This is a flow chart of the optimal band combination algorithm of the present invention; The process of constructing different types of oil film thickness prediction models based on the fluorescence spectrum index of the sample oil film is as follows: S41: For each type of TBFI fluorescence index, all bands within the fluorescence spectrum were traversed. A total of 330 bands were collected during the experiment. The TBFI value of each band combination is calculated, and the TBFI value of each band combination is subjected to partial least square regression with the oil film thickness, and finally the determination coefficient of each band combination is calculated. S42: Determination coefficient R of all band combinations 2 Arrange in descending order and select the band combination with the largest determination coefficient; S43: Combine the band combination with the largest determination coefficient R2 with partial least squares regression PLSR to establish a TBFI-PLSR model, determine the independent variable group (TBFI) and the dependent variable group (oil film thickness), extract the first principal component of the two variable groups, establish the regression of the two variable groups on the principal component, replace the residual matrix, repeat the above steps S41-S43, obtain the regression equation, perform cross-validation, and predict the oil film thickness.

[0026] Based on the model evaluation index, by comparing the evaluation index of four TBFI-PLSR models, the suitable fluorescence index type and the corresponding optimal waveband combination for oil film thickness detection are selected, and the TBFI-PLSR model suitable for oil film thickness detection of gasoline and diesel and the corresponding optimal emission waveband combination are obtained. Based on the TBFI method, the optimal excitation wavelength and emission waveband combination of different thickness oils can be determined in a similar way, so as to realize the detection of oil film thickness; The evaluation index includes the determination coefficient, the root mean square error and the mean square relative error.

[0027]

[0028]

[0029] In the formula, n is the sample number, is the actual observation value (true value) of the i-th sample, is the model prediction value of the i-th sample, is the average value of all actual observation values.

[0030] The calculation formula of the determination coefficient is as follows:

[0031] In the formula, n is the sample number, is the actual observation value (true value) of the i-th sample, is the model prediction value of the i-th sample, is the average value of all actual observation values.

[0032] Embodiment 1: A sea surface oil film thickness detection method based on laser fluorescence spectrum and multi-waveband fluorescence index, comprising the following steps: Experimental preparation, sample preparation First, configure the oil film thickness samples of 92# gasoline and 0# diesel oil with different oil film thicknesses. According to the size of the glassware and the volume of the oil sample, oil film samples with a thickness range of 7.96 μm-795.77 μm are prepared. The experiment is carried out in a dark room to avoid external light interference. The experimental device is as shown in Figure 1 The excitation light source is reflected into the center of the oil film sample through a plane mirror, and the plane mirror is placed at an angle of 45°. The fluorescence signal is collected through an optical fiber and a ground object spectrometer.

[0033] 2. Data acquisition Laser (excitation wavelength is 355 nm) and portable ground object spectrometer are used to collect fluorescence spectrum data of oil film, a total of 330 wavebands. The average value of every 10 spectra is taken as the original fluorescence spectrum of the oil film sample.

[0034] 3. Data preprocessing The raw data were smoothed and filtered to remove noise interference, and the characteristic spectral band within the range of 370-700 nm was selected as the fluorescence spectrum data for subsequent analysis.

[0035] 4. Calculation of Fluorescence Index For gasoline and diesel thickness samples, the following formulas are used to calculate four different types of fluorescence spectral indices:

[0036]

[0037]

[0038]

[0039] in , , Indicates the emission wavelength in the 370-700nm spectral range a , b and c The corresponding fluorescence intensity ( ), EX The excitation wavelength is 355 nm.

[0040] Combine Figure 3 , according to the optimal transmission band combination algorithm: (1) For each TBFI fluorescence index type, all bands within the fluorescence spectrum were traversed. A total of 330 bands were collected during the experiment. The TBFI value of each band combination is calculated, and the TBFI value of each band combination is subjected to partial least squares regression (PLSR) with the oil film thickness, and finally the determination coefficient of each band combination is calculated. (2) The coefficient of determination (R) of all band combinations 2 ) are arranged in descending order and the coefficient of determination (R 2 )The largest band combination; (3) Finally, the coefficient of determination (R 2 ) is combined with partial least squares regression (PLSR) to establish the TBFI-PLSR model to predict the oil film thickness.

[0041] Furthermore, based on the model's evaluation metrics, a TBFI-PLSR model and the corresponding optimal emission band combinations for gasoline and diesel oil film thickness detection were derived. The TBFI method can similarly determine the optimal excitation wavelength and emission band combinations for oils of varying thicknesses, enabling oil film thickness detection.

[0042] TBFI-PLSR based gasoline film thickness prediction results

[0043] Figure 4 Prediction results for 92# gasoline, where (a): scatter plot of measured and predicted values of gasoline film thickness based on TBFI1-PLSR model, (b): scatter plot of measured and predicted values of gasoline film thickness based on TBFI2-PLSR model, (c): scatter plot of measured and predicted values of gasoline film thickness based on TBFI3-PLSR model and (d): scatter plot of measured and predicted values of gasoline film thickness based on TBFI4-PLSR model; Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for detecting sea surface oil film thickness based on laser fluorescence spectroscopy and multi-band fluorescence index, characterized by: The following steps are involved: Obtain fluorescence spectrum data of the oil film sample to be tested, and take the average value of every N spectra as the original fluorescence spectrum of the oil film sample; Preprocess the raw fluorescence spectrum data; For the samples to be tested, the fluorescence spectrum index of the sample oil film is calculated based on different types of fluorescence index calculation formulas; Based on the fluorescence spectrum index of the oil film of the detected sample, different types of oil film thickness prediction models are constructed; Based on the evaluation index, the optimal emission band combination corresponding to different types of oil film thickness prediction models is determined, thereby realizing the detection of oil film thickness.

2. The method for detecting sea surface oil film thickness based on laser fluorescence spectroscopy and multi-band fluorescence index according to claim 1, characterized in that: The calculation formulas for the different types of fluorescence indices are as follows: in: , , Indicates the fluorescence intensity corresponding to emission wavelengths a, b and c in the 370-700nm spectral range, , EX is the excitation wavelength of 355 nm.

3. The method for detecting sea surface oil film thickness based on laser fluorescence spectroscopy and multi-band fluorescence index according to claim 1, characterized in that: The process of constructing different types of oil film thickness prediction models based on the fluorescence spectrum index of the sample oil film is as follows: S41: For each type of TBFI fluorescence index, all bands within the fluorescence spectrum are traversed to calculate the TBFI value under each band combination, and the TBFI value under each band combination is subjected to partial least squares regression with the oil film thickness, and finally the determination coefficient under each band combination is calculated; S42: Arrange the determination coefficients of all band combinations in descending order, and select the band combination with the largest determination coefficient; S43: Combine the band combination with the largest determination coefficient with partial least squares regression to establish a TBFI-PLSR model: determine the independent variable group (TBFI) and the dependent variable group (oil film thickness), extract the first principal component of the two variable groups, establish the regression of the two variable groups on the principal component, replace the residual matrix, repeat the above steps S41-S43, obtain the regression equation, perform cross-validation, and predict the oil film thickness.

4. The method for detecting sea surface oil film thickness based on laser fluorescence spectroscopy and multi-band fluorescence index according to claim 1, characterized in that: The evaluation indicators include determination coefficient, root mean square error and root mean square relative error.

5. The method for detecting sea surface oil film thickness based on laser fluorescence spectroscopy and multi-band fluorescence index according to claim 1, characterized in that: The calculation formula of the determination coefficient is as follows: Where n is the number of samples, is the actual observed value of the i-th sample, i.e., the true value. The model prediction value of the i-th sample, is the average of all actual observations.

6. A device for detecting sea surface oil film thickness based on laser fluorescence spectroscopy and multi-band fluorescence index, characterized by: include: Laser: As an excitation light source, used to illuminate the sample to stimulate fluorescence signals; Power supply: used to power the laser; Spectrometer: used to collect fluorescence spectrum signals; PC: used to display the fluorescence spectrum signal collected by the laser; The PC is connected to the spectrometer via an RJ-45 interface. Detection module: used to receive the spectrum signal collected by the spectrometer and detect the thickness of the sea surface oil film based on the method described in claims 1-5.

Images