55 results about "Oil type" patented technology

The invention discloses a follow-up water drive potential tapping method based on flow field comprehensive analysis, and belongs to the technical field of potential tapping of remaining oil in a high-water-content stage in the middle and later periods of oilfield development. The method comprises the following steps: quantitatively identifying a dominant seepage channel based on streamline distribution, an oil saturation field and a water passing multiple of a target oil reservoir; dividing six remaining oil types and determining distribution characteristics of the remaining oil types; pressure system research and reasonable injection-production ratio regulation and control are combined, and differential potential tapping countermeasures for different remaining oil types are provided. According to the method, the development problems of high water content and low recovery degree in the subsequent water flooding stage of polymer flooding are effectively solved, and the oil reservoir recovery efficiency and development benefits can be remarkably improved.

The invention belongs to the technical field of oil level monitoring of transformer oil conservators, and discloses an oil level monitoring device of an oil conservator, a using method and the oil conservator, the monitoring device comprises a corrugated core body arranged in the oil conservator, a monitoring stand pipe is arranged beside the corrugated core body, the edge of the top of the corrugated core body is connected with a magnetic ring, and the magnetic ring is arranged outside the monitoring stand pipe in a sleeving mode; a monitoring circuit is arranged in the monitoring riser and comprises a power supply, a resistor string and a magnetically controlled switch, one end of the resistor string is connected with the power supply, and the other end is connected with an output end; the resistor string comprises a plurality of resistors which are connected in series; the output end of each resistor is connected with the output end through a magnetically controlled switch; the magnetic control switches are arranged at different heights in the monitoring vertical pipe in the vertical direction. The output end is connected with a resistance measuring device. The technical problem that oil level monitoring is inconvenient due to the fact that the oil conservator is affected by the installation height is solved, and real-time monitoring of the oil level of the corrugated inner oil type oil conservator is achieved.

Composition (C1) for hand disinfection comprising: a) At least one alcohol having one to four carbon atoms, b) At least one water-in-oil type emulsion (E) comprising: a cross-linked anionic polyelectrolyte (P); a fat phase consisting of at least one oil (H), water, a water-in-oil type emulsifying system (S1), and an oil-in-water type emulsifying system (S2); (c) at least one compound of formula (II): R2-(C=O)-O-(CH2-CH2-O)x-CH2-CH[(O-CH2-CH2)yO-(C=O)-R3]-CH2-(O-CH2-CH2)zO-(C=O)-R4 (II) wherein R2-(C=O), R3-(C=O) and R4-(C=O), whether identical or different, represent an acyl group, saturated or unsaturated, linear or branched, comprising from sixteen to twenty-four carbon atoms, and wherein the sum x + y + z is greater than or equal to thirty and less than or equal to one hundred; and (d) water.

To provide a water-in-oil cosmetic that offers both excellent usability as a cosmetic product and high stability. [Solution] A water-in-oil cosmetic composition containing 50% by mass or more of water, comprising the following components (A) to (C): (A) dipolyhydroxy fatty acid polyalkylene glycol, (B) sesquiisostearate sorbitan, and (C) oil thickener. Preferably, component (A) is polyethylene glycol dipolyhydroxystearate, component (B) is sorbitan sesquiisostearate, and component (C) is dextrin palmitate.

This invention provides a method for selecting a three-phase emulsifier that has stable emulsifying ability for the oil type to be emulsified, a three-phase emulsifier, a three-phase emulsion using the emulsifier, and a three-phase emulsification method. [Solution] The method for selecting an emulsifier for three-phase emulsification is to select an emulsifier whose light scattering intensity at 25°C is 1500 or more when dispersed at a concentration of 2% by mass in an oily substance to be emulsified in three phases.

This invention provides an intelligent identification method for residual oil occurrence types in viscosity-assisted steam flooding, comprising: Step 1, preparing core samples and equipment for a viscosity-assisted steam flooding simulation experiment; Step 2, scanning the core samples using CT scanning technology to obtain CT scan image data; Step 3, thresholding the CT scan image data to identify each phase fluid; Step 4, calibrating the residual oil type based on image data from a typical sample set; Step 5, performing convolutional neural network identification of the residual oil type; Step 6, classifying and statistically analyzing the proportion of residual oil types; and Step 7, evaluating the development effect based on the proportion of residual oil types. This intelligent identification method for residual oil occurrence types in viscosity-assisted steam flooding can improve computational efficiency while obtaining higher accuracy and reliability of the residual oil types obtained after viscosity-assisted steam flooding, and more comprehensively evaluating its development effect.

The invention discloses an integrated portable probe system for on-site rapid diagnosis of an oil state, and belongs to the technical field of oil monitoring. The system comprises a portable host box body, an oil path sampling and detecting unit, a data acquisition and processing module, a human-computer interaction interface and a thermal printer module. Wherein the oil path sampling and detecting unit comprises a front-end sampling probe and a rear-end detecting assembly arranged in a box body, is integrated with a micro pump, a cleanliness sensor and a multi-parameter integrated sensor, and can synchronously detect the granularity, moisture, viscosity and acid value of oil liquid. And the data acquisition and processing module performs fusion analysis on the data by adopting a differential weight algorithm based on preset equipment and oil product types, and outputs a comprehensive evaluation report in real time through screen display and a printer. According to the invention, the problems of long detection period, high cost and single function of portable equipment in the prior art are solved, multi-parameter synchronous detection and on-site instant diagnosis are realized, and the operation and maintenance efficiency of the equipment is remarkably improved.

This invention relates to an oil mixing interface detector, belonging to the field of oil mixing interface detection technology. The oil mixing interface detector includes a fixed tube with a protective sleeve fixedly connected to its side wall. A control module is located at the top of the fixed tube, and a detection component is located at the bottom of the fixed tube. An exhaust component is located at the bottom of the detection component. Connecting components are respectively located on the side wall of the exhaust component and the side wall of the protective sleeve. By setting the connecting components and fine-tuning the flexible tube port, the two connecting components can be connected to the input and output pipes respectively. The detection component effectively improves the detection range of multiple oil types, enhances the reliability of the detection, and makes the measurement results more realistic, accurate, and stable. The exhaust component effectively prevents oil overflow, thereby venting the gas in the oil and avoiding any impact on the detection of oil waste heat.

The invention relates to the technical field of preparation of intelligent composite anticorrosive coatings, in particular to 8-hydroxyquinoline-polyglycolic acid microspheres, a preparation method and an anticorrosive coating. The method comprises the following steps: jointly dissolving polyglycolic acid and 8-hydroxyquinoline in hexafluoroisopropanol to obtain a water-phase solution; mixing liquid paraffin and an emulsifier to obtain an oil phase solution; dropwise adding the water-phase solution into the oil-phase solution to form water-in-oil type emulsion droplets, carrying out emulsion polymerization reaction in the water-in-oil type emulsion droplets to obtain 8-hydroxyquinoline-polyglycolic acid microspheres with corrosion inhibition performance and fluorescence indication function, and mixing the 8-hydroxyquinoline-polyglycolic acid microspheres with epoxy resin and a curing agent to prepare the anticorrosive paint. The corrosion inhibition performance and the fluorescence indication effect of 8HQ are utilized, and the thermoplasticity and the fluidity of PGA are utilized, so that the anticorrosive coating is endowed with the self-repairing capability, the problems of unstable performance and single function are solved, and the novel intelligent anticorrosive coating integrating the functions of corrosion protection, in-situ monitoring and self-repairing is realized.

The invention relates to the technical field of fire-retardant hydraulic fluid treatment. The invention relates to an on-line state diagnosis and emergency treatment system for flame-retardant hydraulic fluid. The system comprises a state acquisition unit, a node acquisition unit, a prediction node analysis unit, a node abnormity analysis unit and an emergency processing unit. The state acquisition unit is used for acquiring the performance state of the flame-retardant hydraulic fluid and recording the historical working condition environment and the historical maintenance record of the flame-retardant hydraulic fluid; by dividing a center node and an online node, an abnormal boundary threshold matrix adaptive to different oil types is constructed, and standardized evaluation of the oil state is realized; according to the method, a CNN-LSTM mixed deep learning model is combined, time sequence correlation between spatial features of a working condition environment and maintenance records is fully excavated, short-term, medium-term and long-term prediction node sets are generated, passive monitoring is converted into active pre-judgment, fault early warning time is greatly advanced, and sufficient windows are reserved for operation and maintenance disposal.

The utility model discloses an internal oil type oil conservator with an internal exhaust function, which comprises a conservator body bottom plate, the conservator body bottom plate is used as a supporting foundation of the device, two groups of corrugated core bodies are arranged on the upper surface of the conservator body bottom plate, and each group of corrugated core bodies is provided with a group of first exhaust pipes which are arranged inside each group of corrugated core bodies. One end of the first exhaust pipe penetrates through the bottom plate of the corrugated core body and the bottom plate of the cabinet body to be communicated with the gas collection box and is used for transmitting gas, and the gas collection box is arranged below the bottom plate of the cabinet body and is used for collecting the gas from the first exhaust pipe. According to the gas collecting device, gas can be collected in time when generated, leakage and waste of the gas are avoided, meanwhile, the two first exhaust pipes gather the gas into the gas collecting box, then the gas is discharged through the second exhaust pipe in a unified mode, and the gas discharging efficiency and orderliness are improved.

The application belongs to the technical field of electric power robots. A transformer oil sample collection pressure dynamic adjustment control method and system are proposed. The flow characteristics are analyzed. Based on the real-time pressure and the environmental pressure difference, the oil density and the valve coefficient are combined to calculate the discharge flow. The feedforward-feedback composite control is executed. The pressure deviation is calculated and the instantaneous flow is estimated. The valve new opening degree instruction is output by using the adaptive control law containing the feedforward compensation term. The pressure mutation is actively inhibited. The multi-dimensional parameter online adaptive scheduling is implemented. The PID gain is dynamically adjusted according to the oil temperature and the oil type. The effective volume elastic modulus and the oil density are corrected. The application solves the problems of response lag and poor parameter adaptability of the traditional control, significantly improves the safety of the sampling process, and avoids the equipment damage caused by pressure impact.

The invention discloses an offshore water oil concentration monitoring method based on three-dimensional fluorescence and liquid level compensation. The method comprises the following steps: S1, measuring the liquid level height, the fluorescence intensity and the fluorescence spectrum of a current to-be-measured water body to obtain the real-time liquid level height, the real-time fluorescence intensity and the mixed fluorescence spectrum; s2, acquiring a correction factor under the real-time liquid level height according to a pre-calibrated correction curve, and calculating to obtain corrected fluorescence intensity by combining the real-time fluorescence intensity; s3, comparing the fluorescence spectrum in the step S1 with standard oil type spectrum data to identify the type of the mixed oil type in the water body to be detected; and S4, selecting a corresponding quantitative analysis method according to the spectrum overlapping degree of different oil types in the fluorescence spectrum so as to determine the actual concentration of the mixed oil type. According to the invention, the influence of liquid level fluctuation on light intensity is dynamically offset, and the accuracy and reliability of concentration measurement are ensured; by utilizing a three-dimensional fluorescence spectrum and a linear superposition principle, identification and accurate quantification of mixed oil types are realized.

The invention discloses an oil film bidirectional reflection characteristic on-site measurement device and an oil film thickness inversion correction method, and belongs to the technical field of image processing, and the oil film bidirectional reflection characteristic on-site measurement device comprises an experiment pool which contains an oil-water mixed sample; the annular guide rail is arranged around the experimental water tank; the camera support is arranged on the annular guide rail in a sliding mode, and a fixing base capable of rotating around a horizontal shaft and a vertical shaft is arranged at the upper end of the camera support; the hyperspectral camera is fixed on the fixed seat and is used for measuring the radiance of the experimental pool; and the irradiance meter is used for measuring the irradiance of the sunlight. The oil film bidirectional reflection characteristic on-site measurement device is simpler in structure and easy to implement, and the spectrum difference caused by different visual angles of the sensor is eliminated. And the accuracy and reliability of oil type identification and thickness inversion are greatly improved.

A gearbox lubricating and flushing system relates to the technical field of transmission equipment. The system comprises a gearbox, a lubricating oil tank and a flushing oil tank, lubricating oil is stored in the lubricating oil tank, a lubricating circulating oil way is formed between the lubricating oil tank and the gearbox through a first pipeline, and a lubricating adjusting assembly is arranged on the lubricating circulating oil way and used for driving the lubricating oil to circulate and adjusting the pressure of the lubricating oil in real time. Flushing oil is stored in the flushing oil tank, a flushing circulating oil way is formed between the flushing oil tank and the gear box through a second pipeline, a flushing adjusting assembly is arranged on the flushing circulating oil way, and the lubricating circulating oil way and the flushing circulating oil way are mutually independent, can conduct lubricating and cleaning respectively and have independent lubricating and flushing functions. The oil type, temperature and pressure during lubrication and flushing can be independently controlled according to test requirements, so that the parameters are matched with the test requirements, and the test requirements are met.

The utility model particularly discloses an oil leakage prevention type ventilation cap which comprises a column body, external threads are arranged at the bottom end of the column body, the interior of the column body is of a through hole structure, a filter element is arranged in the through hole structure, and a top cover is arranged at the top of the column body. The filter element made of the super-strong oleophylic and hydrophobic materials is installed in the lubricating oil filter, residual oil mist can be effectively adsorbed, meanwhile, water vapor in air can be prevented from permeating, and the service life of lubricating oil is effectively prolonged.

The invention relates to the technical field of cosmetic anticorrosion, and particularly discloses a mild anticorrosion composition and a preparation method and application thereof.The composition is prepared from, by mass, 13.0%-16.7% of oil-soluble alcohol, 2.0%-5.0% of fatty acid salt, 1.0%-3.0% of plant extract, 1%-3% of polyol ester and the balance antibacterial grease; the fatty acid salt is zinc octoate and sodium lauroyl sarcosinate; the plant extract is a moutan bark extract and carnosic acid; the components take effect synergistically to achieve long-acting antibacterial and antiseptic effects, the components are mild and non-irritant, synergistic bacteriostasis is achieved through multiple mechanisms of pH adjustment, microbial cell membrane damage, metabolic enzyme activity inhibition and the like, the oil solubility is excellent, and the composition is suitable for pure oil agents, water-in-oil type cream and oil-in-water type cream. The problems that a traditional preservative is strong in irritation, poor in oil phase dispersion and difficult to pass the international standard test are solved, and the preservative has the characteristics of mildness, high efficiency and low addition amount.

The invention relates to a low-temperature quick starting method for an electronic oil pump, and aims to solve the technical problem that the rotating speed response speed of an existing electronic oil pump is insufficient under the working conditions of low temperature and load change along with time. The method comprises the following steps of: starting at the maximum phase current when judging that the actual rotating speed is lower than a preset entering power limit rotating speed threshold in the initial starting stage, and ensuring the starting efficiency; when the actual rotating speed is higher than the threshold value, power limiting closed-loop control is started, the increasing and decreasing step length of a reference rotating speed instruction is dynamically adjusted by comparing the current power with the rated power, and the step length can be adaptively adjusted through a table look-up method according to the oil viscosity coefficient, the oil temperature and the real-time rotating speed; and finally, realizing closed-loop control of the motor through PI control, coordinate transformation and an SVPWM algorithm based on the reference rotating speed instruction. The electronic oil pump can be rapidly started at low temperature and rapidly responded to the rotating speed, meanwhile, the function safety of the oil pump is guaranteed, and the electronic oil pump is suitable for different oil products and working condition changes and wide in application range.

The invention relates to the technical field of oil product doping detection, and discloses an oil product detection model generation method and device, equipment and a medium, and the method comprises the steps: obtaining a spectral signal data set of a to-be-detected oil product; constructing a complete feature vector of the discrete channel based on the derivative data features and the basic signal of the preset dimension; performing multi-round iterative processing on the complete feature vector based on a preset prediction model, and determining an oil product prediction result output by the current round based on an oil product prediction result output by the previous round and a fitting residual error generated by the current round; and obtaining a final prediction result of the complete feature vector of the discrete channel after multiple rounds of iteration processing, determining a prediction deviation between the final prediction result and an actual oil product type of the oil product to be detected, and constructing an optimization objective function according to the prediction deviation and a regularization item matched with the prediction model, and the prediction model is optimized by optimizing the objective function. The oil product detection device has the beneficial effect of improving the oil product detection precision.

The present application relates to oil leakage detection technical field, disclose a kind of booster station transformer oil leakage detection method, device, equipment and storage medium, in the method provided by the present application, fusion data is constructed using hyperspectral data, visible light image and temperature data, and the denoising and dimensionality reduction of hyperspectral data and visible light image data are realized using image preprocessing method, then the region segmentation model after training is used to segment the oil leakage area to be verified, to realize the accurate division of oil leakage area, then combined with component recognition model and regional statistical analysis method, identify the oil type in the oil leakage area to be verified, simultaneously quantitatively obtain oil film thickness and temperature distribution data, multidimensional accurate extraction of oil leakage characteristics of the oil leakage area to be verified;Finally, by comprehensive oil type, temperature distribution and oil film thickness output as oil leakage detection result, realize timely and accurate detection of the oil leakage condition of booster station transformer.

The utility model discloses a hot oil type energy-saving heating roller, which belongs to the field of printing machine accessories and comprises a roller body, the middle of the roller body is hollow to form a cavity, two sealing blocks are arranged in the roller body in a sliding mode, oil conveying holes are arranged in the middles of the sealing blocks, one end of each oil conveying hole is communicated with the cavity, and the other end of each oil conveying hole is connected with a heating oil pump through an oil conveying pipe. By means of the inner cavity of the heating roller, heated hot oil is injected, heat of the hot oil is transmitted to the heating roller, the heating effect is achieved, the contact area of the inner cavity and the hot oil can be adjusted through the sealing baffle, and therefore the contact position of the hot oil and the heating roller is adjusted, the heating position can be adjusted according to needs, and the heating effect is improved. According to the utility model, the positions which do not work cannot be heated, so that the situations that the whole heating roller is large in heating area and needs more energy investment due to heating of the whole heating roller, and the energy is flat and lost due to the fact that the heating positions do not need to be put into production are reduced.

The application discloses a shipborne radar oil spill detection method and system based on an improved whale optimization algorithm, and relates to the technical field of image processing and pattern recognition.The technical points of the application include: preprocessing the obtained shipborne radar image;extracting the saliency boundary feature of the preprocessed image;extracting the ROI image based on the saliency boundary feature;obtaining the optimal segmentation threshold value by using the improved whale optimization algorithm, and segmenting the ROI image by using the optimal segmentation threshold value to obtain the optimal segmentation image;and post-processing the optimal segmentation image to obtain the final oil film image.The application highlights the difference between the oil spill area and the background by extracting the saliency feature;the improved whale optimization algorithm enhances the local search ability of the algorithm and avoids premature convergence.The application realizes efficient and accurate identification of the oil film area, can reduce the false detection rate, and is suitable for detection scenes of different sea conditions and oil spill types.

The invention discloses a product oil demand sensitivity prediction method and device. The method comprises the following steps: acquiring statistical data of product oil; wherein the statistical data comprises historical product oil sales volume data and historical product oil price change parameters; carrying out quantity-price relation modeling on the historical product oil sales volume data and the historical product oil price change parameters to obtain a product oil sales volume prediction model; determining the demand sensitivity of the product oil through the product oil sales prediction model; according to the method, statistical data in a plurality of oil price adjustment periods are fully utilized, and the difference of gasoline and diesel oil in market response characteristics is combined to construct adaptive sales prediction models, so that the sales change trend of different product oil types under different price adjustment amplitudes can be accurately reflected, and the sales prediction accuracy is improved. The subjectivity and the uncertainty brought by a traditional artificial experience judgment or static average estimation method are effectively overcome, and the precision and the practicability of demand sensitivity analysis are remarkably improved.

The invention relates to a lubricating oil stagnation point selling method and system based on distributed storage, and the method comprises the following steps: 110, obtaining a lubricating oil filling request of a user, and obtaining the real-time position of the user when the lubricating oil filling request exists; 120, pushing an optimal lubricating oil stagnation point in the lubricating oil selling network to the user based on the real-time position of the user, and generating a navigation path from the real-time position to the optimal lubricating oil stagnation point; step 130, acquiring a lubricating oil type selected by the user in the optimal lubricating oil stagnation point; step 140, based on the types of the lubricating oil, lubricating oil is filled for the user through lubricating oil equipment, and the flow of the filled lubricating oil is read after filling is completed; 150, based on the lubricating oil flow, cost accounting is conducted according to a preset price, and the cost is sent to a user for payment. According to the invention, related problems of lubricating oil selling in an existing business mode can be well solved, a proper lubricating oil stagnation point can be conveniently found for a user, and the consumption experience of the user is optimized.

Provided is a water-in-oil type emulsion cosmetic composition which has excellent anti-caking properties of ultraviolet scattering agent particles, excellent emulsion stability and usability, and excellent skin fusion properties after application. A water-in-oil type emulsion cosmetic composition comprising an oily medium, ultraviolet scattering agent particles dispersed in the oily medium, a dispersant for the ultraviolet scattering agent particles, a silicone-based surfactant having an HLB value of 8.0 or less, and water, condition (I): the oily medium contains a polar oil and / or a polar silicone oil, and a non-polar silicone oil, or contains a polar oil and / or a polar silicone oil, a hydrocarbon oil, and a non-polar silicone oil, and at least a portion of the oily medium is a volatile component; condition (II): the amount of the nonpolar silicone oil is 55 parts by mass or more with respect to 100 parts by mass of the total of the oily medium; (III) the dispersant contains polyhydroxystearic acid and / or polyglycerol polyricinoleate, and the content ratio of the ultraviolet scattering agent particles to the dispersant is 10.0 or less; condition (IV): the content of the silicone-based surfactant is 3.0 mass% or less. And condition (V): the content ratio of the silicone surfactant to the dispersant is less than 1.0.

This invention belongs to the field of environmental emergency technology, specifically disclosing a method for analyzing the treatment effect of oily wastewater based on image recognition. It addresses the problems of traditional methods such as infrared and ultraviolet methods for measuring oil content in water, which involve numerous instruments, cumbersome operations, long processing times, and poor adaptability to various scenarios. This invention combines deep learning-based target detection technology with microscopic images of oily wastewater. A target detection model is used to learn and train on the number, volume, and color of emulsified oil droplets in wastewater with different oil contents, treatment processes, and oil types. In field applications, the target detection model is used to identify and calculate the microscopic images of oily wastewater, thereby quickly predicting the oil content. This method is highly operable, accurate, and enables real-time analysis of the treatment effect of oily wastewater, while also being highly adaptable to changes in application scenarios.

The invention provides a Pickering type sunscreen emulsion and a preparation method thereof. The Pickering type sunscreen emulsion provided by the invention is a water-in-oil type sunscreen emulsion and comprises an oil phase and a water phase, the oil phase contains modified nanoparticles; the modified nanoparticles are inorganic nanoparticles modified by fatty acid, alkyl phosphate or alkyl alkoxy silane; the inorganic nanoparticles comprise at least one of zinc oxide and titanium dioxide; the mass of the modified nanoparticles is 16%-30% of the mass of the oil phase; the volume ratio of the water phase to the oil phase is (3-7): (3-7). Amphiphilic modified nanoparticles are introduced into an oil phase, and the addition amount and the water-oil volume ratio of the modified nanoparticles are controlled, so that the modified nanoparticles can form a compact and high-strength particle interface film on an oil-water interface, and finally the Pickering emulsion with excellent stability is obtained; in addition, inorganic nano-particles in the modified nano-particles are inorganic physical sun-screening agents, and irritation is low.

The application discloses a microcosmic residual oil type conversion quantitative analysis method based on CT technology. The method comprises the following steps: collecting a core sample of a to-be-detected area, simulating a displacement process of the core sample, and collecting two-dimensional CT scanning images corresponding to the core sample under at least two different displacement states in the displacement process; for the two-dimensional CT scanning images under each displacement state, an oil phase three-dimensional model under the displacement state is obtained by recombination respectively; according to a three-dimensional shape factor G of a residual oil cluster in the oil phase three-dimensional model and an Euler number E of a pore space N , a microcosmic occurrence type of the residual oil cluster is divided; according to a conversion relationship of the residual oil cluster, a microcosmic residual oil type conversion three-dimensional model is obtained. The microcosmic residual oil type conversion three-dimensional model is applied to calculate a residual oil cluster conversion rate and an actual production proportion. The application is more accurate in residual oil type division and quantitative analysis, and is beneficial to improve the production degree of microcosmic residual oil.

The utility model discloses an observation window structure and an oil type transformer, relates to the related technical field of oil type transformers, and is used for solving the problem that the observation process of the oil quantity and the residual space in an oil collecting cavity of an oil collecting part is tedious. The observation window structure is applied to the oil collecting piece, an oil collecting cavity used for collecting oil is formed in the oil collecting piece, an opening communicated with the oil collecting cavity is formed in a top cover of the oil collecting piece, and the observation window structure is used for being arranged at the opening; the observation window structure comprises a body and an oil leakage hole formed in the body, the body is provided with a first surface and a second surface which are arranged oppositely, and the oil leakage hole penetrates through the first surface and the second surface. The residual capacity in the oil collection cavity can be rapidly and visually observed through the oil leakage hole, and the effect of collecting leaked oil is achieved.

The present application relates to the technical field of marine oil spill monitoring, in particular to a marine oil spill monitoring method and system based on multi-band fluorescence spectrum neural network. The method comprises the following steps: acquiring fluorescence spectrum and fluorescence image data of different oil types and different oil film thicknesses of oil spills; preprocessing the fluorescence spectrum and fluorescence image data of marine oil spills respectively; constructing three-dimensional fluorescence spectrum data and thickness difference matrix by using the preprocessed fluorescence spectrum; extracting band features and thickness features of the three-dimensional fluorescence spectrum data and thickness difference matrix by using a deep neural network; the present application effectively integrates marine oil spill fluorescence spectrum and fluorescence image data, and provides a reliable basis for accurate identification and early warning of oil spills.