Crude oil spectrum-based marine oil spill monitoring method and system

Abstract

The invention relates to the technical field of marine environment monitoring, in particular to a marine oil spill monitoring method and system based on a crude oil spectrum. Comprising the following steps: S1, identifying N marine oil spill positions by using a fuzzy reasoning model, and distributing monitoring points for the N marine oil spill positions to form a first monitoring sequence; s2, K important oil spill positions and corresponding monitoring points are optimized through a genetic algorithm, a second monitoring sequence is formed, and K is smaller than N and is a positive integer larger than or equal to 1; s3, calculating the time when the crude oil reaches the sea surface and marking a path; a high-resolution spectrometer is used to collect spectral data in visible light, near-infrared and mid-infrared bands. Marine oil spill is accurately positioned through fuzzy reasoning and a genetic algorithm, a high-resolution spectrometer and a Transform model are utilized to predict spectral data, a monitoring path is fused and optimized, energy loss and solar spectrum influence are considered, the model and the path are continuously corrected, and efficient and accurate monitoring is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of marine environment monitoring, and in particular to a method and system for monitoring marine oil spills based on crude oil spectrum. Background Art

[0002] Crude oil is an important part of the global energy system and plays an indispensable role in economic development, industrial production, transportation and daily life. However, marine oil spills can cause serious damage to the marine ecosystem, pollute water quality, poison marine life, affect fisheries and tourism, and lead to economic losses and ecological disasters. Existing marine oil spill monitoring methods, such as satellite remote sensing, drone monitoring and surface ship inspections, have the advantages of wide coverage and strong real-time performance, but they also have the disadvantages of high cost, great weather influence and limited resolution. In order to improve the accuracy and reliability of monitoring, spectral technology plays an important role in marine oil spill monitoring. It can provide high-precision oil pollution identification and classification, and can quickly detect the composition and concentration of oil pollution, thereby improving the efficiency of emergency response. However, despite the introduction of spectral technology, the existing marine oil spill monitoring methods still have deficiencies in resolution, false alarm rate and the ability to distinguish oil pollution from other substances, resulting in the need to further improve the accuracy and reliability of monitoring results. Summary of the invention

[0003] In order to overcome the shortcoming of low accuracy of marine oil spill monitoring, the present invention provides a marine oil spill monitoring method and system based on crude oil spectrum.

[0004] The technical implementation scheme of the present invention is: a method for monitoring marine oil spills based on crude oil spectrum, comprising the following steps: S1: using the fuzzy inference model to identify N marine oil spill locations, and assigning monitoring points to the N marine oil spill locations to form a first monitoring sequence; S2: Optimize K important oil spill locations and their corresponding monitoring points through a genetic algorithm to form a second monitoring sequence, where K is less than N and is a positive integer greater than or equal to 1; S3: Calculate the time when the crude oil reaches the sea surface and mark the path; use a high-resolution spectrometer to collect spectral data in the visible light, near-infrared and mid-infrared bands, and screen features after preprocessing; use the Transformer prediction model to predict the spectral data, fuse the first and second monitoring sequences, and form the first fused monitoring path; S4: Considering the influence of energy loss and solar spectrum, the first fusion monitoring path is adjusted to form a second fusion monitoring path; the actual value of the crude oil spectrum data actually monitored is compared with the predicted value, and the difference is obtained to correct the prediction model and path; S5: Mark the oil spill location according to the difference, and focus on the marked area; the whole process is repeated to continuously optimize the monitoring effect.

[0005] Preferably, the method of using a fuzzy inference model to identify N marine oil spill locations and assigning monitoring points to the N marine oil spill locations to form a first monitoring sequence includes: collecting and preprocessing marine platform, submarine pipeline and marine environment data, defining fuzzy set membership functions, and establishing a fuzzy rule base; performing fuzzy inference, including data fuzzification, rule matching and inference results, and finally defuzzifying and converting the results into specific marine oil spill locations; According to the location of the marine oil spill, the sea surface projection point corresponding to the marine oil spill location is taken as the center, and the center is the marine monitoring location; the farthest sea surface monitoring range is allocated to the N marine monitoring locations, and then the N marine monitoring locations and the farthest sea surface monitoring ranges allocated to the N marine monitoring locations constitute the first marine oil spill monitoring sequence.

[0006] Preferably, the method of optimizing K important oil spill locations and their corresponding monitoring points by genetic algorithm to form a second monitoring sequence, where K is less than N and K is a positive integer greater than or equal to 1, includes: generating an initial set of potential marine oil spill locations, evaluating the fitness of each location, and selecting locations with high fitness to enter the next generation; performing crossover and mutation operations to generate a new set of marine oil spill locations; merging the parent generation and the offspring generation, selecting marine oil spill locations with high fitness to enter the next generation, until the maximum fitness threshold is reached, and selecting the optimal K marine oil spill locations; allocating the optimal K marine oil spill locations to K marine monitoring locations, and forming a second marine oil spill monitoring sequence according to the first monitoring sequence construction rules.

[0007] Preferably, the calculation of the time when the crude oil reaches the sea surface and marking the path includes: obtaining depth data of the location of the marine oil spill, and dividing the marine depth data into three stages according to the maximum influencing factor: depth affected by ocean current, depth affected by wind, and depth affected by water temperature; for each depth stage, dividing the depth data under the maximum influencing factor of the stage by the influence coefficient of the maximum influencing factor on the rising speed of the crude oil, calculating the rising time of the crude oil spill in the stage, and obtaining the crude oil rising time of the three depth stages; based on the crude oil rising time of the three stages, obtaining the time when the crude oil spill reaches the sea surface, marking each medium path for the crude oil spill to reach the sea surface, and obtaining a path marking sequence, wherein the path marking sequence content includes the maximum influencing factor, depth, and rising time; allocating a path marking sequence to the first monitoring position and the second monitoring position, and defining them as the first path marking sequence and the second path marking sequence, respectively.

[0008] Preferably, the method of assigning a path marking sequence to the first monitoring position and the second monitoring position, and defining them as the first path marking sequence and the second path marking sequence respectively, comprises: based on the path marking sequence, extracting time data in each path marking sequence, the time data comprising the rising time of three stages; firstly, sorting the different oil spill positions as a whole according to the total time for the crude oil to reach the sea surface from the spill position, that is, sorting them from long to short according to the total time; then, sorting the three-stage depth time of the different oil spill positions, the specific steps are as follows: sorting the time at the depth affected by the current from long to short; sorting the time at the depth affected by the wind from long to short; sorting the time at the depth affected by the water temperature from long to short.

[0009] Preferably, the method of predicting spectral data using a Transformer prediction model includes: obtaining a training set of crude oil spectral data in a first path label sequence and a second path label sequence, constructing a Transformer prediction model, inputting the training set data into the Transformer prediction model to obtain a trained Transformer prediction model, and using the Transformer prediction model to obtain predicted values ​​of the crude oil spectral data in the first path label sequence and the second path label sequence.

[0010] Preferably, the fusion of the first and second monitoring sequences to form a first fused monitoring path includes: sorting the predicted values ​​of the sea surface crude oil spectral data from large to small to determine the monitoring priority; obtaining the sorting results in the first path marking sequence and the second path marking sequence, and the construction method of the fused monitoring sequence is as follows: the longest overall time: select the oil spill location with the longest total time; the longest depth time of the three stages: when the total time is the same, ensure that the depth time of the three stages is also the longest.

[0011] Preferably, the energy loss and the influence of the solar spectrum are taken into consideration to adjust the first fused monitoring path to form a second fused monitoring path; the actual value of the crude oil spectrum data monitored is compared with the predicted value to obtain the difference for correcting the prediction model and path, including: based on the fused monitoring path, evaluating the negative impact of the energy loss and solar spectrum in each monitoring path on the field monitoring results; according to the evaluation results, sorting according to the size of the negative impact, that is, sorting from large to small according to the negative impact; adjusting the monitoring path according to the sorting result to form a new fused monitoring path.

[0012] Preferably, the monitoring degree of the oil spill location is marked according to the size of the difference, with a focus on the marked area, including: based on the crude oil spectrum difference data, recording the number of corrections of the predicted value of the sea surface crude oil spectrum data and the second fusion monitoring path; using the number of corrections as the first weight of the monitoring degree mark, and using the crude oil spectrum difference size as the second weight of the monitoring degree mark; determining the priority monitoring position based on the weight marking result, and the specific method is to sort from large to small according to the total weight, and select the position with the highest weight as the priority monitoring position; then build the entire system into a circulation mechanism, regularly update the monitoring path and adjust the weight to ensure continuous optimization of the monitoring effect.

[0013] A marine oil spill monitoring system based on crude oil spectrum, comprising: Fuzzy reasoning and genetic algorithm module: collect and preprocess ocean-related data, use fuzzy reasoning models to determine the location of oil spills, and use genetic algorithms to optimize the location of monitoring points to improve monitoring efficiency and accuracy; Path analysis module: Calculate the time it takes for crude oil to reach the sea surface based on depth data and mark each oil spill path; assign path marking sequences according to monitoring points to form an orderly monitoring path; Spectral data acquisition and processing module: Use a high-resolution spectrometer to collect spectral data of the ocean surface within a specific wavelength range, perform preprocessing, use fuzzy reasoning to screen key spectral features, and predict spectral data based on the Transformer model; Fusion monitoring path module: combines the spectral data prediction values ​​in different monitoring sequences to form a comprehensive monitoring path, connecting all oil spill locations to ensure full coverage of potential oil spill areas; Energy loss and solar spectrum impact module: Consider the impact of energy loss and solar spectrum on monitoring results, adjust the monitoring path, reduce negative impact, and optimize monitoring results; Monitoring optimization and feedback module: compares the difference between the predicted value and the actual monitoring value, modifies the Transformer prediction model and monitoring path according to the difference, marks the key monitoring locations, and optimizes the entire monitoring process in a loop to continuously improve the accuracy and response speed of the system.

[0014] Beneficial effects: The present invention realizes efficient marine oil spill monitoring and management through a multi-step comprehensive technical means. First, the fuzzy reasoning model and genetic algorithm are used to accurately determine and optimize the marine oil spill location and monitoring points, thereby improving the accuracy and efficiency of monitoring. Then, by calculating the time when the crude oil reaches the sea surface and marking the path, the comprehensive coverage and real-time tracking of the oil spill path are ensured. Then, a high-resolution spectrometer is used to collect and process the spectral data of the ocean surface, and the predicted value of the spectral data is obtained using the Transformer prediction model, which further improves the accuracy and reliability of monitoring. Subsequently, different monitoring sequences are fused to form a comprehensive monitoring path, ensuring comprehensive coverage of potential oil spill areas. In addition, considering the influence of energy loss and solar spectrum, the monitoring path is adjusted to reduce the negative impact of external factors on the monitoring results. Finally, by comparing the difference between the predicted value and the actual monitoring value, the prediction model and monitoring path are continuously corrected, the key monitoring location is marked, and the monitoring effect is continuously optimized through a loop mechanism to ensure the accuracy and response speed of the system. In summary, the present invention not only improves the accuracy, efficiency and comprehensiveness of marine oil spill monitoring, but also ensures long-term monitoring effects through continuous optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A flow chart of a method for monitoring marine oil spills based on crude oil spectrum of the present invention; Figure 2 The schematic diagram of the structure of a marine oil spill monitoring system based on crude oil spectrum of the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] Example 1: A method for monitoring marine oil spills based on crude oil spectra, such as Figure 1 As shown, the following steps are included: S1: using the fuzzy inference model to identify N marine oil spill locations, and assigning monitoring points to the N marine oil spill locations to form a first monitoring sequence; S2: Optimize K important oil spill locations and their corresponding monitoring points through a genetic algorithm to form a second monitoring sequence, where K is less than N and is a positive integer greater than or equal to 1; S3: Calculate the time when the crude oil reaches the sea surface and mark the path; use a high-resolution spectrometer to collect spectral data in the visible light, near-infrared and mid-infrared bands, and screen features after preprocessing; use the Transformer prediction model to predict the spectral data, fuse the first and second monitoring sequences, and form the first fused monitoring path; S4: Considering the influence of energy loss and solar spectrum, the first fusion monitoring path is adjusted to form a second fusion monitoring path; the actual value of the crude oil spectrum data actually monitored is compared with the predicted value, and the difference is obtained to correct the prediction model and path; S5: Mark the oil spill location according to the difference, with a focus on the marked area; the whole process is repeated to continuously optimize the monitoring effect.

[0018] Using the fuzzy reasoning model to identify N marine oil spill locations, and assigning monitoring points to the N marine oil spill locations, forming a first monitoring sequence, including: collecting and preprocessing marine platform, submarine pipeline and marine environmental data, defining fuzzy set membership functions, and establishing a fuzzy rule base; performing fuzzy reasoning, including data fuzzification, rule matching and reasoning results, and finally defuzzifying and converting the results into specific marine oil spill locations; According to the location of the marine oil spill, the sea surface projection point corresponding to the marine oil spill location is taken as the center, and the center is the marine monitoring location; the farthest sea surface monitoring range is allocated to the N marine monitoring locations, and then the N marine monitoring locations and the farthest sea surface monitoring ranges allocated to the N marine monitoring locations constitute the first marine oil spill monitoring sequence.

[0019] Further explanation is, fuzzy reasoning model: a method based on fuzzy logic to deal with uncertainty and imprecise data to infer the location of marine oil spills by defining membership functions and rule bases; Assuming that the coordinates of the oil spill location are (10, 20), with this coordinate as the center, the farthest monitoring position on the sea surface with a radius of 5 kilometers is allocated; the specific monitoring positions include (10, 25), (15, 20), (5, 20) and (10, 15), which constitute the first marine oil spill monitoring sequence: [(10, 25, 5km), (15, 20, 5km), (5, 20, 5km), (10, 15, 5km)], and the sequence includes the central coordinates of each monitoring position and the farthest monitoring range.

[0020] K important oil spill locations and their corresponding monitoring points are optimized by genetic algorithm to form a second monitoring sequence, where K is less than N and K is a positive integer greater than or equal to 1, including: generating an initial set of potential marine oil spill locations, evaluating the fitness of each location, and selecting locations with high fitness to enter the next generation; performing crossover and mutation operations to generate a new set of marine oil spill locations; merging the parent generation and the offspring generation, selecting marine oil spill locations with high fitness to enter the next generation until the maximum fitness threshold is reached, and selecting the optimal K marine oil spill locations; allocating the optimal K marine oil spill locations to K marine monitoring locations, and forming a second marine oil spill monitoring sequence according to the first monitoring sequence construction rules.

[0021] A further explanation is that the genetic algorithm is an optimization technique that imitates the natural selection process and searches for the optimal solution through selection, crossover and mutation operations to optimize the location of monitoring points. It is assumed that an initial set of potential oil spill locations is generated, such as (10, 20), (15, 25), (20, 30). The fitness of each location is evaluated and locations with high fitness are selected to enter the next generation. Crossover and mutation operations are performed to generate a new set of locations. The parent and child generations are merged and locations with high fitness are selected to enter the next generation until the maximum fitness threshold is reached and the optimal K locations are selected. The K locations are assigned corresponding marine monitoring locations to form the second marine oil spill monitoring sequence: [(17, 27, 5km), (19, 28, 5km)], where each location includes the center coordinates and the farthest monitoring range. The farthest monitoring range of the sea surface is the maximum sea area that can be effectively monitored by each monitoring location, which is used to determine the layout of the monitoring network.

[0022] The method calculates the time when crude oil reaches the sea surface and marks the path, including: obtaining depth data of the location of the marine oil spill, and dividing the marine depth data into three stages according to the maximum influencing factor, namely, depth affected by ocean current, depth affected by wind, and depth affected by water temperature according to the depth data; for each depth stage, dividing the depth data under the maximum influencing factor of the stage by the influence coefficient of the maximum influencing factor on the rising speed of crude oil, calculating the rising time of the crude oil spill in the stage, and obtaining the crude oil rising time of the three depth stages; based on the crude oil rising time of the three stages, obtaining the time when the crude oil spill reaches the sea surface, marking each medium path for the crude oil spill to reach the sea surface, and obtaining a path marking sequence, wherein the path marking sequence includes the maximum influencing factor, depth, and rising time; allocating a path marking sequence to a first monitoring position and a second monitoring position, and defining them as a first path marking sequence and a second path marking sequence, respectively.

[0023] Further explanation is that by considering the influence of various factors on the rising speed of crude oil at different depths, the accuracy of time prediction is improved; the path mark sequence records the maximum influencing factors, depth and rising time of the three stages, which is helpful to track the movement path of crude oil, provide a scientific basis for subsequent monitoring and response measures, and ensure timely and effective oil spill treatment; the path mark sequence: a series of marks with timestamps and other relevant information (such as depth, influencing factors), indicating the movement path of crude oil from the leakage point to the sea surface; Assuming that the depth of the oil spill location is 100 meters, it is divided into three stages: depth affected by ocean currents, depth affected by wind force, and depth affected by water temperature; the rising time of the three stages is calculated: the ocean current impact stage is 1 hour, the wind impact stage is 2 hours, and the water temperature impact stage is 1.5 hours; the total time is 4.5 hours; the path marking sequence is: [(ocean current, 30m, 1h), (wind force, 70m, 2h), (water temperature, 100m, 1.5h)], which records the maximum influencing factors, depths and rising times of the three stages.

[0024] A path marking sequence is allocated to a first monitoring position and a second monitoring position, and they are defined as a first path marking sequence and a second path marking sequence respectively, including: based on the path marking sequence, extracting time data in each path marking sequence, the time data including the rising time of three stages; firstly, sorting the different oil spill positions as a whole according to the total time for the crude oil to reach the sea surface from the spill position, that is, sorting them from long to short according to the total time; then, sorting the three-stage depth time of different oil spill positions, the specific steps are as follows: sorting the time at the depth affected by the current from long to short; sorting the time at the depth affected by the wind from long to short; sorting the time at the depth affected by the water temperature from long to short.

[0025] For further explanation, suppose there are three different ocean oil spill locations, each with a corresponding path marker sequence: Ocean oil spill location A: path marker sequence: [(ocean current, 30m, 1h), (wind force, 70m, 2h), (water temperature, 100m, 1.5h)], Ocean oil spill location B: path marker sequence: [(ocean current, 20m, 0.5h), (wind force, 60m, 1.8h), (water temperature, 90m, 1.2h)], Ocean Oil spill location C: path mark sequence: [(ocean current, 25m, 0.8h), (wind force, 65m, 1.6h), (water temperature, 95m, 1.4h)]; extract the time data in each path mark sequence: ocean oil spill location A: [1h, 2h, 1.5h], ocean oil spill location B: [0.5h, 1.8h, 1.2h], ocean oil spill location C: [0.8h, 1.6h, 1.4h]; put all the time data together for Overall sorting: Overall time data: [1h, 2h, 1.5h, 0.5h, 1.8h, 1.2h, 0.8h, 1.6h, 1.4h], overall sorting result: [0.5h, 0.8h, 1h, 1.2h, 1.4h, 1.5h, 1.6h, 1.8h, 2h]. Through overall sorting, the total time for the crude oil to reach the sea surface at each oil spill location is determined: Marine oil spill location A: 1h+2h+1.5h=4.5h, Marine Oil spill location B: 0.5h+1.8h+1.2h=3.5h, ocean oil spill location C: 0.8h+1.6h+1.4h=3.8h. The final sorting is based on the total time to sort the different oil spill locations: Ocean oil spill location A: 4.5h, Ocean oil spill location C: 3.8h, Ocean oil spill location B: 3.5h. In this way, priority is given to monitoring the oil spill locations where the crude oil takes the longest time to reach the sea surface, ensuring the effective use of resources and timely response. Sorting by stage depth time, time sorting at the depth affected by ocean current: time data: [1h, 0.5h, 0.8h], sorting result: [1h, 0.8h, 0.5h], time sorting at the depth affected by wind: time data: [2h, 1.8h, 1.6h], sorting result: [2h, 1.8h, 1.6h], time sorting at the depth affected by water temperature: time data: [1.5h, 1.2h, 1.4h], sorting result: [1.5h, 1.4h, 1.2h]. By sorting by stage depth time, the impact degree of the three stages can be clarified, and the monitoring path and resource allocation can be optimized; for example, for the depth affected by ocean current, the time at oil spill location A is the longest and should be monitored first; similarly, for the depth affected by wind and water temperature, the priority monitoring locations should be determined.

[0026] The method uses a Transformer prediction model to predict spectral data, including: obtaining a training set of crude oil spectral data in a first path label sequence and a second path label sequence, constructing a Transformer prediction model, inputting the training set data into the Transformer prediction model to obtain a trained Transformer prediction model, and using the Transformer prediction model to obtain predicted values ​​of the crude oil spectral data in the first path label sequence and the second path label sequence.

[0027] Further explanation is that by processing complex spectral data through deep learning models, it is possible to capture subtle changes in spectral features, improve the accuracy and reliability of predictions, and provide a scientific basis for subsequent monitoring and analysis; assuming that the first path label sequence is [(ocean current, 30m, 1h), (wind force, 70m, 2h)], and the second path label sequence is [(water temperature, 100m, 1.5h), (ocean current, 20m, 0.5h)]; extract the crude oil spectral data training set from these path label sequences, and build and train the Transformer model; the trained model is used to predict the crude oil spectral data in the first path label sequence and the second path label sequence to obtain predicted values, such as [0.8, 0.6] and [0.7, 0.5]; these predicted values ​​help accurately evaluate the composition and state of crude oil; the crude oil spectral data: spectral information reflected or emitted by the ocean surface collected by a high-resolution spectrometer, which reflects the existence and characteristics of crude oil; The meaning of the predicted values: [0.8, 0.6]: These are the predicted values ​​of the crude oil spectral data at two positions in the first path marker sequence; 0.8: represents the predicted value of the crude oil spectral data at the first position; 0.6: represents the prediction of the crude oil spectral data at the second position; [0.7, 0.5]: These are the predicted values ​​of the crude oil spectral data at two positions in the second path marker sequence; 0.7: represents the predicted value of the crude oil spectral data at the first position; 0.5: represents the predicted value of the crude oil spectral data at the second position; These predicted values ​​help to accurately assess the composition and state of crude oil, and provide a scientific basis for subsequent monitoring and analysis; for example, the predicted values ​​are used to determine the type, concentration, and degree of contamination of crude oil, thereby guiding emergency response and environmental protection measures.

[0028] The first and second monitoring sequences are integrated to form a first integrated monitoring path, including: sorting the predicted values ​​of the sea surface crude oil spectrum data from large to small to determine the monitoring priority; obtaining the sorting results in the first path marker sequence and the second path marker sequence, and the construction method of the fused monitoring sequence is as follows: the longest overall time: select the oil spill location with the longest total time; the longest depth time in the three stages: when the total time is the same, ensure that the depth time of the three stages is also the longest.

[0029] Further explanation is, assuming that the first path label sequence is [(ocean current, 30m, 1h), (wind force, 70m, 2h)], and the second path label sequence is [(water temperature, 100m, 1.5h), (ocean current, 20m, 0.5h)]; extract crude oil spectral data from these path label sequences, train the Transformer model, and obtain the predicted values: the first path label sequence is [0.8, 0.6], and the second path label sequence is [0.7, 0.5]; 1. Sort the predicted values ​​of the sea surface crude oil spectral data from large to small: the sorted predicted values: [0.8, 0.7, 0.6, 0.5], 2. Obtain the sorting results in the first path label sequence and the second path label sequence, and the construction method of the fused monitoring sequence is as follows: the first path label sequence: [0.8, 0.6], the second path label sequence: [0.7, 0.5], 3. According to the overall sorting result and the depth sorting results of each stage, the first Monitoring sequence and the second monitoring sequence: Overall sorting results: [(ocean current, 30m, 1h, 0.8), (water temperature, 100m, 1.5h, 0.7), (wind force, 70m, 2h, 0.6), (ocean current, 20m, 0.5h, 0.5)], Time sorting under the depth affected by the ocean current: [(ocean current, 30m, 1h, 0.8), (ocean current, 20m, 0.5h, 0.5)], Time sorting under the depth affected by the wind: [(wind force, 70m , 2h, 0.6)],Time sorting under water temperature influence depth: [(water temperature, 100m, 1.5h, 0.7)],4. Form a fusion monitoring sequence: [(ocean current, 30m, 1h, 0.8), (water temperature, 100m, 1.5h, 0.7), (wind force, 70m, 2h, 0.6), (ocean current, 20m, 0.5h, 0.5)],In this way, it ensures that the priority and distribution of monitoring points are more reasonable, and improves the monitoring efficiency and accuracy.

[0030] Taking into account the influence of energy loss and solar spectrum, the first fused monitoring path is adjusted to form a second fused monitoring path; the actual value of the crude oil spectrum data actually monitored is compared with the predicted value, and the difference is obtained to correct the prediction model and path, including: based on the fused monitoring path, evaluating the negative impact of energy loss and solar spectrum in each monitoring path on the field monitoring results; according to the evaluation results, sorting according to the size of the negative impact, that is, sorting from large to small; adjusting the monitoring path according to the sorting result to form a new fused monitoring path.

[0031] Further explanation is that the energy loss and solar spectrum data: the energy loss caused by environmental factors (such as lighting conditions) during the monitoring process and the impact of sunlight on the spectral data, these data are used to adjust the monitoring strategy to reduce interference; the specific evaluation method includes experimental data or simulation results, and the evaluation indicators include energy consumption and the degree of reduction in monitoring accuracy; the specific adjustment method includes optimizing the path sequence and path layout to reduce energy loss and improve monitoring accuracy; assuming that the fused monitoring path includes four locations: [(ocean current, 30m, 1h, 0.8), (water temperature, 100m, 1.5h, 0.7), (wind force, 70m, 2h, 0.6), (ocean current, 20m, 0.5h, 0.5)]; according to the negative impact of energy loss and solar spectrum in each monitoring path on the field monitoring results, The impact values ​​are: [0.9, 0.7, 0.8, 0.6]; 1. Sort by impact from large to small: the sorted path: [(ocean current, 30m, 1h, 0.8, 0.9), (wind force, 70m, 2h, 0.6, 0.8), (water temperature, 100m, 1.5h, 0.7, 0.7), (ocean current, 20m, 0.5h, 0.5, 0.6)], 2. Form a new fusion monitoring path: [(ocean current, 30m, 1h, 0.8, 0.9), (wind force, 70m, 2h, 0.6, 0.8), (water temperature, 100m, 1.5h, 0.7, 0.7), (ocean current, 20m, 0.5h, 0.5, 0.6)], in this way, the monitoring path is optimized, the energy loss and the negative impact of the solar spectrum are reduced, and the accuracy and efficiency of monitoring are improved.

[0032] The oil spill location is marked for monitoring degree according to the size of the difference, with a focus on the marked area, including: based on the crude oil spectrum difference data, recording the number of corrections between the predicted value of the sea surface crude oil spectrum data and the second fusion monitoring path; using the number of corrections as the first weight of the monitoring degree mark, and the size of the crude oil spectrum difference as the second weight of the monitoring degree mark; determining the priority monitoring location based on the weight marking results, and specifically sorting from large to small according to the total weight, and selecting the location with the highest weight as the priority monitoring location; then building the entire system into a circulation mechanism, regularly updating the monitoring path and adjusting the weight to ensure continuous optimization of the monitoring effect.

[0033] Further explanation is that the monitoring degree mark: the importance level assigned to each monitoring position according to the difference in crude oil spectrum is used to determine the key areas for priority monitoring; the first weight, the more corrections, the greater the weight; the second weight, the greater the difference, the greater the weight; the specific recording method uses a database or log file; assuming that the second fusion monitoring path includes four positions: [(ocean current, 30m, 1h, 0.8), (water temperature, 100m, 1.5h, 0.7), (wind , 70m, 2h, 0.6), (ocean current, 20m, 0.5h, 0.5)]; According to the field monitoring data, calculate the crude oil spectrum difference data: [0.1, 0.2, 0.05, 0.15]; 1. Record the number of corrections: [(ocean current, 30m, 1h, 0.8, 2 times), (water temperature, 100m, 1.5h, 0.7, 1 time), (wind force, 70m, 2h, 0.6, 3 times), (ocean current, 20m, 0.5h, 0.5, 1 time)], 2. Determine the weight Weight: The number of corrections is used as the first weight: [2, 1, 3, 1], the difference in crude oil spectrum is used as the second weight: [0.1, 0.2, 0.05, 0.15], 3. Calculate the total weight: [(ocean current, 30m, 1h, 0.8, 2, 0.1, 2.1), (water temperature, 100m, 1.5h, 0.7, 1, 0.2, 1.2), (wind force, 70m, 2h, 0.6, 3, 0.05, 3.05), (ocean current, 20m, 0.5h, 0.5, 1, 0.15, 1.15)], 4. Determine the priority monitoring locations: Sort by total weight: [(wind force, 70m, 2h, 0.6, 3, 0.05, 3.05), (ocean current, 30m, 1h, 0.8, 2, 0.1, 2.1), (water temperature, 100m, 1.5h, 0.7, 1, 0.2, 1.2), (ocean current, 20m, 0.5h, 0.5, 1, 0.15, 1.15)], In this way, determine the priority monitoring locations and continuously optimize the monitoring effect.

[0034] Embodiment 2: Based on Embodiment 1, a marine oil spill monitoring system based on crude oil spectrum includes:

[0035] Fuzzy reasoning and genetic algorithm module: collect and preprocess ocean-related data, use fuzzy reasoning models to determine the location of oil spills, and use genetic algorithms to optimize the location of monitoring points to improve monitoring efficiency and accuracy; Path analysis module: Calculate the time it takes for crude oil to reach the sea surface based on depth data and mark each oil spill path; assign path marking sequences according to monitoring points to form an orderly monitoring path; Spectral data acquisition and processing module: Use a high-resolution spectrometer to collect spectral data of the ocean surface within a specific wavelength range, perform preprocessing, use fuzzy reasoning to screen key spectral features, and predict spectral data based on the Transformer model; Fusion monitoring path module: combines the spectral data prediction values ​​in different monitoring sequences to form a comprehensive monitoring path, connecting all oil spill locations to ensure full coverage of potential oil spill areas; Energy loss and solar spectrum impact module: Consider the impact of energy loss and solar spectrum on monitoring results, adjust the monitoring path, reduce negative impact, and optimize monitoring results; Monitoring optimization and feedback module: compares the difference between the predicted value and the actual monitoring value, modifies the Transformer prediction model and monitoring path according to the difference, marks the key monitoring locations, and optimizes the entire monitoring process in a loop to continuously improve the accuracy and response speed of the system.

[0036] The above is a detailed introduction to the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for monitoring marine oil spills based on crude oil spectrum, characterized in that: The following steps are involved: S1: using the fuzzy inference model to identify N marine oil spill locations, and assigning monitoring points to the N marine oil spill locations to form a first monitoring sequence; S2: Optimize K important oil spill locations and their corresponding monitoring points through a genetic algorithm to form a second monitoring sequence, where K is less than N and is a positive integer greater than or equal to 1; S3: Calculate the time when the crude oil reaches the sea surface and mark the path; use a high-resolution spectrometer to collect spectral data in the visible light, near infrared and mid-infrared bands, and screen features after preprocessing; The spectral data is predicted using the Transformer prediction model, and the first and second monitoring sequences are fused to form a first fused monitoring path; S4: Considering the influence of energy loss and solar spectrum, the first fusion monitoring path is adjusted to form a second fusion monitoring path; the actual value of the crude oil spectrum data actually monitored is compared with the predicted value, and the difference is obtained to correct the prediction model and path; S5: Mark the oil spill location according to the difference, with a focus on the marked area; the whole process is repeated to continuously optimize the monitoring effect.

2. A method for monitoring marine oil spills based on crude oil spectrum according to claim 1, characterized in that: The method uses a fuzzy inference model to identify N marine oil spill locations and allocates monitoring points to the N marine oil spill locations to form a first monitoring sequence, including: collecting and preprocessing marine platform, submarine pipeline and marine environment data, defining fuzzy set membership functions, and establishing a fuzzy rule base; performing fuzzy inference, including data fuzzification, rule matching and inference results, and finally defuzzifying and converting the results into specific marine oil spill locations; According to the location of the marine oil spill, the sea surface projection point corresponding to the marine oil spill location is taken as the center, and the center is the marine monitoring location; the farthest sea surface monitoring range is allocated to the N marine monitoring locations, and then the N marine monitoring locations and the farthest sea surface monitoring ranges allocated to the N marine monitoring locations constitute the first marine oil spill monitoring sequence.

3. A method for monitoring marine oil spills based on crude oil spectrum according to claim 1, characterized in that: The method optimizes K important oil spill locations and their corresponding monitoring points through a genetic algorithm to form a second monitoring sequence, where K is less than N and K is a positive integer greater than or equal to 1, and includes: generating an initial set of potential marine oil spill locations, evaluating the fitness of each location, and selecting locations with high fitness to enter the next generation; performing crossover and mutation operations to generate a new set of marine oil spill locations; merging the parent generation and the offspring generation, selecting marine oil spill locations with high fitness to enter the next generation, until a maximum fitness threshold is reached, and selecting the optimal K marine oil spill locations; allocating the optimal K marine oil spill locations to K marine monitoring locations, and forming a second marine oil spill monitoring sequence according to the first monitoring sequence construction rule.

4. The method for monitoring marine oil spills based on crude oil spectrum according to claim 1, characterized in that: The method calculates the time when crude oil reaches the sea surface and marks the path, including: obtaining depth data of the location of the marine oil spill, and dividing the marine depth data into three stages according to the maximum influencing factor, namely, depth affected by ocean current, depth affected by wind, and depth affected by water temperature according to the depth data; for each depth stage, dividing the depth data under the maximum influencing factor of the stage by the influence coefficient of the maximum influencing factor on the rising speed of crude oil, calculating the rising time of the crude oil spill in the stage, and obtaining the crude oil rising time of the three depth stages; based on the crude oil rising time of the three stages, obtaining the time when the crude oil spill reaches the sea surface, marking each medium path for the crude oil spill to reach the sea surface, and obtaining a path marking sequence, wherein the path marking sequence includes the maximum influencing factor, depth, and rising time; and allocating the path marking sequence to the first monitoring position and the second monitoring position, and defining them as the first path marking sequence and the second path marking sequence, respectively.

5. A method for monitoring marine oil spills based on crude oil spectrum according to claim 4, characterized in that: The method allocates a path mark sequence to the first monitoring position and the second monitoring position, and defines them as the first path mark sequence and the second path mark sequence respectively, including: extracting time data in each path mark sequence based on the path mark sequence, the time data including the rise time of three stages; firstly, sorting the different oil spill positions as a whole according to the total time for the crude oil to reach the sea surface from the spill position, that is, sorting them from long to short according to the total time; then, sorting the three-stage depth time of the different oil spill positions, and the specific steps are as follows: sorting the time at the depth affected by the ocean current from long to short; sorting the time at the depth affected by the wind from long to short; and sorting the time at the depth affected by the water temperature from long to short.

6. The method for monitoring marine oil spills based on crude oil spectrum according to claim 1, characterized in that: The method of predicting spectral data using a Transformer prediction model includes: obtaining a training set of crude oil spectral data in a first path label sequence and a second path label sequence, constructing a Transformer prediction model, inputting the training set data into the Transformer prediction model to obtain a trained Transformer prediction model, and using the Transformer prediction model to obtain predicted values ​​of the crude oil spectral data in the first path label sequence and the second path label sequence.

7. The method for monitoring marine oil spills based on crude oil spectrum according to claim 1, characterized in that: The fusion of the first and second monitoring sequences to form a first fused monitoring path includes: sorting the predicted values ​​of the sea surface crude oil spectrum data from large to small to determine the monitoring priority; obtaining the sorting results in the first path mark sequence and the second path mark sequence, and the construction method of the fused monitoring sequence is as follows: the longest overall time: selecting the oil spill location with the longest total time; the longest depth time in the three stages: when the total time is the same, ensure that the depth time of the three stages is also the longest.

8. The method for monitoring marine oil spills based on crude oil spectrum according to claim 1, characterized in that: The first fusion monitoring path is adjusted to form a second fusion monitoring path by considering the influence of energy loss and solar spectrum; Compare the actual value of the crude oil spectrum data monitored with the predicted value, and obtain the difference to correct the prediction model and path, including: based on the fusion monitoring path, evaluate the negative impact of energy loss and solar spectrum in each monitoring path on the field monitoring results; According to the evaluation results, the negative impacts are sorted according to their magnitude, that is, they are sorted from large to small; the monitoring path is adjusted according to the sorting results to form a new integrated monitoring path.

9. The method for monitoring marine oil spills based on crude oil spectrum according to claim 1, characterized in that: The method marks the monitoring degree of the oil spill location according to the size of the difference, focusing on the marked area, including: based on the crude oil spectrum difference data, recording the number of corrections of the predicted value of the sea surface crude oil spectrum data and the second fusion monitoring path; using the number of corrections as the first weight of the monitoring degree mark, and using the size of the crude oil spectrum difference as the second weight of the monitoring degree mark; determining the priority monitoring position according to the weight marking result, and specifically sorting from large to small according to the total weight, and selecting the position with the highest weight as the priority monitoring position; then building the entire system into a circulation mechanism, regularly updating the monitoring path and adjusting the weight to ensure continuous optimization of the monitoring effect.

10. A marine oil spill monitoring system based on crude oil spectrum, used in the marine oil spill monitoring method based on crude oil spectrum according to any one of claims 1 to 9, characterized in that: Included are: Fuzzy reasoning and genetic algorithm module: collect and preprocess ocean-related data, use fuzzy reasoning models to determine the location of oil spills, and use genetic algorithms to optimize the location of monitoring points to improve monitoring efficiency and accuracy; Path analysis module: Calculate the time it takes for crude oil to reach the sea surface based on depth data and mark each oil spill path; assign path marking sequences according to monitoring points to form an orderly monitoring path; Spectral data acquisition and processing module: Use a high-resolution spectrometer to collect spectral data of the ocean surface within a specific wavelength range, perform preprocessing, use fuzzy reasoning to screen key spectral features, and predict spectral data based on the Transformer model; Fusion monitoring path module: combines the spectral data prediction values ​​in different monitoring sequences to form a comprehensive monitoring path, connecting all oil spill locations to ensure full coverage of potential oil spill areas; Energy loss and solar spectrum impact module: Consider the impact of energy loss and solar spectrum on monitoring results, adjust the monitoring path, reduce negative impact, and optimize monitoring results; Monitoring optimization and feedback module: compares the difference between the predicted value and the actual monitoring value, modifies the Transformer prediction model and monitoring path according to the difference, marks the key monitoring locations, and cyclically optimizes the entire monitoring process to continuously improve the accuracy and response speed of the system.

Images