Monitoring and early warning method and system for marine three-dimensional layered rights allocation projects

Abstract

This invention discloses a method and system for monitoring and early warning of marine three-dimensional hierarchical rights-granting projects, belonging to the field of marine monitoring and early warning technology. The method continuously acquires time-series marine monitoring datasets within a designated marine oil extraction area and performs feature analysis to obtain a set of monitoring and early warning indices for each time point within the designated marine oil extraction area, including a sea surface extraction stress index, a mid-ocean strata operation disturbance index, and a seabed extraction comprehensive risk index. A comprehensive analysis of the monitoring and early warning index sets for each time point within the designated marine oil extraction area yields an oil extraction early warning index for each time point. This invention improves the accuracy of early warning by providing intelligent oil extraction early warning based on the oil extraction early warning index for each time point within the designated marine oil extraction area, thereby timely identifying potential risks and providing strong protection for the safety of marine extraction activities.

Description

Technical Field

[0001] This invention relates to the field of marine monitoring and early warning technology, specifically to a method and system for monitoring and early warning of marine three-dimensional hierarchical weighted projects. Background Technology

[0002] With the development of the global economy and technology, the development and utilization of marine resources has become increasingly important, especially in the fields of energy, mineral resource extraction, and marine ecological protection. While activities such as deep-sea oil and gas extraction and seabed mining bring huge economic benefits, they also pose potential threats to the marine environment. In order to ensure the efficient development of marine resources and the sustainable development of marine ecology, marine environmental monitoring and early warning systems have become crucial technical means. The proposed three-dimensional marine layered monitoring technology aims to detect potential risks in the marine environment in a timely manner through comprehensive monitoring of various layers of the ocean. It can comprehensively and in real time acquire multi-dimensional marine environmental data, provide scientific basis for decision-makers, and ensure the safety of marine resource development and the effectiveness of environmental protection.

[0003] Existing technology, such as the patent application with publication number CN119479211A, discloses a marine environmental monitoring and early warning method and system based on big data. The steps are as follows: This method, relating to the field of marine environmental monitoring technology, includes: collecting environmental data in real time through sensors within monitoring stations and transmitting it to a database; using positioning technology to obtain the location information of each monitoring station and calculating the distance between a randomly selected benchmark monitoring station and its corresponding comparison monitoring station; analyzing the changing trends of environmental data and determining the propagation speed of abnormal environmental data based on the analysis results; combining distance and propagation speed, using a linear propagation algorithm to calculate the arrival time of abnormal environmental data from the benchmark monitoring station to the comparison monitoring station; setting the alarm activation time based on the arrival time and issuing alarm information. This invention, through the deployment of monitoring stations and the integration of sensor technology, achieves real-time monitoring and data collection of the marine environment, providing a comprehensive and accurate information foundation for subsequent data analysis and early warning.

[0004] Based on the above findings, the limitations of existing technologies include at least the following problems: Existing technologies lack comprehensive analysis of different layers of the ocean, making it difficult to fully and accurately assess the mutual influence of environmental data at each layer when facing complex changes in the marine environment. For example, changes in surface ocean climate can easily affect mid-level water flow, and changes in mid-level water flow can indirectly affect bottom topography and microbial activity. This results in a lag in the response of early warning systems to changes in the marine environment, failing to capture dynamic changes in the surface, mid-level, and bottom layers of the ocean in a timely manner, thus affecting the accuracy and timeliness of early warning information and reducing the safety of offshore oil extraction. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for monitoring and early warning of marine three-dimensional stratified rights projects. This solves the problem that existing technologies are unable to fully reflect the complex changes in the marine environment, which affects the accuracy and timeliness of early warning information and reduces the safety of offshore oil extraction.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a marine three-dimensional layered weighted project monitoring and early warning method, comprising the following steps: continuously acquiring marine monitoring time-series datasets within a designated marine oil extraction area, including marine surface monitoring time-series data, marine mid-layer monitoring time-series data, and marine bottom monitoring time-series data; performing feature analysis on the marine monitoring time-series datasets within the designated marine oil extraction area to obtain a monitoring and early warning index set for each time point within the designated marine oil extraction area, including a sea surface extraction stress index, a marine mid-layer operational disturbance index, and a seabed extraction comprehensive risk index; comprehensively analyzing the monitoring and early warning index set for each time point within the designated marine oil extraction area to obtain an oil extraction early warning index for each time point within the designated marine oil extraction area; and providing intelligent oil extraction early warning based on the oil extraction early warning index for each time point within the designated marine oil extraction area.

[0007] Furthermore, the specific formula for calculating the oil extraction early warning index at each time point within the designated offshore oil extraction area is as follows: ;in, To define the first [area] within the offshore oil extraction zone Oil extraction early warning index at a specific time point. To define the first [area] within the offshore oil extraction zone The sea surface mining stress index at a given time point The stress coefficient of the sea surface is stored in the database. The stress adjustment coefficient for sea surface data stored in the database. The smoothing coefficients are stored in the database. To define the first [area] within the offshore oil extraction zone Ocean mid-level operational disturbance index at a given time point These are the mid-level perturbation coefficients stored in the database. These are the mid-level perturbation adjustment coefficients stored in the database. To define the first [area] within the offshore oil extraction zone The comprehensive risk index of seabed mining at a given time point. The seabed risk coefficient is stored in the database. The seabed risk adjustment coefficient is stored in the database. The interaction coefficients are stored in the database. , 1, 2, 3, ... , The number of time points.

[0008] Furthermore, the time-series data of ocean surface monitoring includes the sea surface climate stability index, sea surface evaporation rate, wave height, and sea surface photosynthetic intensity at each time point. The specific steps to obtain the sea surface mining stress index at each time point within the designated offshore oil extraction area are as follows: Standardize the sea surface climate stability index, sea surface evaporation rate, wave height, and sea surface photosynthetic intensity at each time point within the designated offshore oil extraction area; and comprehensively analyze the standardized sea surface climate stability index, sea surface evaporation rate, wave height, and sea surface photosynthetic intensity at each time point within the designated offshore oil extraction area to obtain the sea surface mining stress index at each time point within the designated offshore oil extraction area.

[0009] Furthermore, the time-series data of the mid-ocean monitoring includes the mid-layer density jump index, ocean current intensity value, oxygen-sulfur corrosion index, and water shear stress value. The specific steps for obtaining the mid-ocean operational disturbance index at each time point within the designated offshore oil extraction area are as follows: normalize the mid-layer density jump index, ocean current intensity value, oxygen-sulfur corrosion index, and water shear stress value at each time point within the designated offshore oil extraction area; and comprehensively analyze the normalized mid-layer density jump index, ocean current intensity value, oxygen-sulfur corrosion index, and water shear stress value at each time point within the designated offshore oil extraction area to obtain the mid-ocean operational disturbance index at each time point within the designated offshore oil extraction area.

[0010] Furthermore, the specific steps for obtaining the mid-layer density jump index at each time point within the designated offshore oil extraction area are as follows: obtain the temperature and salinity values ​​at several depths at each time point within the designated offshore oil extraction area, and perform comprehensive analysis to obtain the temperature gradient and salinity gradient values ​​at each time point within the designated offshore oil extraction area; and perform comprehensive analysis on the temperature gradient and salinity gradient values ​​at each time point within the designated offshore oil extraction area to obtain the mid-layer density jump index at each time point within the designated offshore oil extraction area.

[0011] Furthermore, the marine bottom monitoring time-series data includes seabed topographic image data, seabed pulsation risk index, and seabed microbial activity index at each time point. Specifically, the seabed topographic image data consists of the pixel value and two-dimensional coordinates of each pixel in the seabed topographic map. The specific steps for obtaining the comprehensive seabed mining risk index at each time point within a designated marine oil extraction area are as follows: input the seabed topographic image data at each time point within the designated marine oil extraction area into a pre-trained topographic recognition model for predictive analysis to obtain the seabed topographic mining risk index at each time point within the designated marine oil extraction area; and comprehensively analyze the seabed topographic mining risk index, seabed disaster index, and seabed microbial activity index at each time point within the designated marine oil extraction area to obtain the comprehensive seabed mining risk index at each time point within the designated marine oil extraction area.

[0012] Furthermore, the specific steps for calculating the comprehensive risk index of seabed extraction at each time point within the designated offshore oil extraction area are as follows: ;in, To define the first [area] within the offshore oil extraction zone The comprehensive risk index of seabed mining at a given time point. To define the first [area] within the offshore oil extraction zone Risk index of seabed topography mining at a given time point. The adjustment coefficients for the seabed topography stored in the database. To define the first [area] within the offshore oil extraction zone The risk index of seabed pulsation at a specific time point. This refers to the pulsation risk adjustment coefficient stored in the database. To define the first [area] within the offshore oil extraction zone Seabed microbial activity index at a specific time point This refers to the adjustment coefficient for microbial activity stored in the database. These are the collaborative adjustment coefficients stored in the database. 1, 2, 3, ... , The number of time points.

[0013] Furthermore, the terrain recognition model specifically uses a convolutional neural network, which includes an input layer, several convolutional layers, activation layers, pooling and flattening layers, and a risk output layer. The specific steps for obtaining the seabed terrain risk index for each time point within a designated offshore oil extraction area are as follows: In the input layer of the convolutional neural network, seabed terrain image data for each time point within the designated offshore oil extraction area is received and preprocessed; in the convolutional layers of the convolutional neural network, feature extraction processing is performed on the preprocessed seabed terrain image data for each time point within the designated offshore oil extraction area to obtain the risk index for each time point within the designated offshore oil extraction area. The process involves generating a seabed topographic feature map. In the activation layer of a convolutional neural network, a nonlinear transformation is performed on the seabed topographic feature map at each time point within a designated offshore oil extraction area. In the pooling and flattening layer of the convolutional neural network, the activated seabed topographic feature map at each time point within the designated offshore oil extraction area is pooled and flattened to obtain a seabed topographic feature vector at each time point within the designated offshore oil extraction area. Finally, in the risk output layer of the convolutional neural network, the seabed topographic feature vectors at each time point within the designated offshore oil extraction area are fused to obtain a seabed topographic extraction risk index at each time point within the designated offshore oil extraction area.

[0014] Furthermore, the specific steps for intelligent oil extraction early warning based on the oil extraction early warning index at each time point within the designated offshore oil extraction area are as follows: Trend analysis is performed on the oil extraction early warning index at each time point within the designated offshore oil extraction area to obtain several sets of oil extraction early warning index change rates within the designated offshore oil extraction area; a comprehensive analysis is then performed on the change rates of each set of oil extraction early warning indices within the designated offshore oil extraction area and the oil extraction early warning index at each time point to obtain the predicted oil extraction early warning index within the designated offshore oil extraction area, which is considered as the oil extraction early warning index for the next time point within the designated offshore oil extraction area; corresponding preset early warning measures are then implemented based on the oil extraction early warning index for the next time point within the designated offshore oil extraction area.

[0015] The marine three-dimensional stratified rights-granting project monitoring and early warning system includes: a marine monitoring time series acquisition module, used to continuously acquire marine monitoring time series datasets within a designated marine oil extraction area, including marine surface monitoring time series data, marine mid-layer monitoring time series data, and marine bottom monitoring time series data; a marine stratification feature analysis module, used to perform feature analysis on the marine monitoring time series datasets within the designated marine oil extraction area to obtain a set of monitoring and early warning indices for each time point within the designated marine oil extraction area, including the sea surface extraction stress index, the marine mid-layer operation disturbance index, and the seabed extraction comprehensive risk index; a comprehensive early warning assessment module, used to comprehensively analyze the monitoring and early warning index set for each time point within the designated marine oil extraction area to obtain an oil extraction early warning index for each time point within the designated marine oil extraction area; and an intelligent early warning module, used to provide intelligent oil extraction early warning based on the oil extraction early warning index for each time point within the designated marine oil extraction area.

[0016] The present invention has the following beneficial effects:

[0017] (1) The marine three-dimensional layered weighted project monitoring and early warning method analyzes the real-time monitoring data of the ocean surface, middle layer and bottom layer in a three-dimensional and hierarchical manner, so as to fully understand the dynamic changes of the marine environment. By comprehensively analyzing the influencing factors at different levels, it provides more accurate early warning for oil extraction. For example, the impact of sea surface climate change on middle layer water flow and the impact of middle layer water flow on bottom layer topography and microbial activity have been effectively assessed, thereby improving the accuracy of early warning, enabling more timely identification of potential risks, and providing strong protection for the safety of marine extraction activities.

[0018] (2) The marine three-dimensional layered rights setting project monitoring and early warning method uses a convolutional neural network to perform in-depth analysis of seabed topographic image data and combines data such as seabed pulsation risk index and microbial activity index to accurately assess the impact of the seabed environment on mining operations. For example, seabed topographic images can reveal potential deep-sea cracks, unstable areas of sediments or caves, which can easily pose a serious threat to equipment such as seabed oil and gas pipelines and drilling platforms. Combined with the pulsation risk index, the impact of seabed vibration on seabed facilities can be assessed. Pulsation changes can easily cause equipment failures or leakage events, thereby enabling early identification of potential threats in the seabed environment, providing more reliable risk warnings for seabed mining, and ensuring the safety and environmental protection of marine oil mining.

[0019] (3) The monitoring and early warning method of the marine three-dimensional layered rights project, by combining data from various layers of the marine environment, especially by analyzing factors such as the intensity of water flow and oxygen-sulfur corrosion index in the middle layer of the ocean, can accurately assess and predict the potential impact of the middle layer environment on oil extraction operations. Disturbances in the middle layer of the water have an important impact on marine oil extraction. For example, strong ocean currents can easily lead to wear or even breakage of oil and gas pipelines. The oxygen-sulfur corrosion index can reflect the concentration of corrosive substances in the water and affect the durability of seabed facilities. By monitoring the middle layer environmental data in real time, the possible environmental risks during the operation can be identified and early warnings can be issued, so as to take effective countermeasures in advance, thereby improving the efficiency and sustainability of the operation.

[0020] (4) The marine three-dimensional layered rights setting project monitoring and early warning system integrates multiple functional modules to achieve efficient acquisition, analysis and early warning of real-time environmental data in the marine oil extraction area. The marine monitoring time sequence acquisition module can continuously monitor the environmental data of the surface, middle and bottom layers of the ocean, thereby ensuring the integrity and timeliness of the data. The marine layered feature analysis module conducts in-depth analysis of data at different levels, thereby accurately assessing the potential impact of each level on oil extraction activities. The intelligent early warning module adjusts the operation strategy in real time based on the oil extraction early warning index generated by the comprehensive early warning assessment module, thereby effectively avoiding potential risks caused by environmental changes and improving the safety of marine oil extraction activities. In turn, it ensures the smooth progress of extraction operations in complex marine environments, while reducing equipment damage and environmental accidents.

[0021] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0022] Figure 1 This is a flowchart of the marine three-dimensional stratified weighted project monitoring and early warning method of the present invention.

[0023] Figure 2 This is a time series diagram of the mid-layer density jump index in the marine oil extraction area set in the marine three-dimensional stratified weighted project monitoring and early warning method of the present invention.

[0024] Figure 3 This invention provides a time series diagram of ocean current intensity within an offshore oil extraction area, used in the marine three-dimensional stratified weighted project monitoring and early warning method.

[0025] Figure 4 This is a time series diagram of the oxygen and sulfur corrosion index in the marine oil extraction area set in the marine three-dimensional layered weighted project monitoring and early warning method of the present invention.

[0026] Figure 5This invention provides a time series diagram of water shear stress within an offshore oil extraction area, used in the marine three-dimensional stratified weighted project monitoring and early warning method.

[0027] Figure 6 This is a flowchart illustrating the specific steps involved in obtaining the comprehensive risk index of seabed extraction at each time point within a designated offshore oil extraction area in the marine three-dimensional stratified weighted project monitoring and early warning method of the present invention.

[0028] Figure 7 This is a block diagram of the marine three-dimensional layered weighted project monitoring and early warning system of the present invention. Detailed Implementation

[0029] See also Figure 1 This invention provides a technical solution: a method for monitoring and early warning of marine three-dimensional layered rights-based projects, comprising the following steps: continuously acquiring marine monitoring time-series datasets within a designated marine oil extraction area, including marine surface monitoring time-series data, marine mid-layer monitoring time-series data, and marine bottom monitoring time-series data; performing feature analysis on the marine monitoring time-series datasets within the designated marine oil extraction area to obtain a monitoring and early warning index set for each time point within the designated marine oil extraction area, including a sea surface extraction stress index (i.e., the impact of the marine surface environment on oil extraction), a marine mid-layer operational disturbance index, and a seabed extraction comprehensive risk index; comprehensively analyzing the monitoring and early warning index set for each time point within the designated marine oil extraction area to obtain an oil extraction early warning index for each time point within the designated marine oil extraction area; and performing intelligent oil extraction early warning based on the oil extraction early warning index for each time point within the designated marine oil extraction area.

[0030] The specific formula for calculating the oil extraction early warning index at each time point within a designated offshore oil extraction area is as follows: ;in, To define the first [area] within the offshore oil extraction zone Oil extraction early warning index at a specific time point. To define the first [area] within the offshore oil extraction zone The sea surface mining stress index at a given time point The stress coefficient of the sea surface is stored in the database. The stress adjustment coefficient for sea surface data stored in the database. This is the smoothing coefficient stored in the database, and in this implementation example, it takes the value of 0.001. To define the first [area] within the offshore oil extraction zone Ocean mid-level operational disturbance index at a given time point These are the mid-level perturbation coefficients stored in the database. These are the mid-level perturbation adjustment coefficients stored in the database. To define the first [area] within the offshore oil extraction zone The comprehensive risk index of seabed mining at a given time point. The seabed risk coefficient is stored in the database. The seabed risk adjustment coefficient is stored in the database. The interaction coefficients are stored in the database. , 1, 2, 3, ... , The number of time points.

[0031] It needs to be explained that in the formula This item is used to adjust the superposition effect among the sea surface mining stress index, the mid-ocean strata operation disturbance index, and the seabed mining comprehensive risk index, so as to avoid the oil extraction early warning index being too high or too low.

[0032] , , The following steps can be used to obtain the following: Read the sea surface mining stress index, mid-ocean strata operation disturbance index, and seabed mining comprehensive risk index for each time point within the designated offshore oil extraction area, and perform summation analysis to obtain the warning sum value. Then, perform a ratio analysis between the sea surface mining stress index, mid-ocean strata operation disturbance index, and seabed mining comprehensive risk index for each time point within the designated offshore oil extraction area and the warning sum value, and use the ratio analysis results as the corresponding coefficients.

[0033] , , , The following steps can be taken to obtain the following: Using historical data, combined with the surface mining stress index, mid-ocean strata disturbance index, and seabed mining comprehensive risk index, statistical regression analysis is performed to quantify the specific impact of each factor on the oil extraction early warning index, thereby fitting initial weight values. Next, sensitivity analysis is used to adjust the value range of each coefficient and observe its impact on the oil extraction early warning results, ensuring the stability and rationality of the model. Based on the characteristics and actual conditions of the oil extraction area, the initially fitted coefficients are corrected and optimized, and finally, coefficient values ​​applicable to specific oil extraction areas are determined.

[0034] Specifically, the time-series data of ocean surface monitoring includes the sea surface climate stability index, sea surface evaporation rate, wave height, and sea surface photosynthetic intensity at each time point. The specific steps to obtain the sea surface extraction stress index at each time point within the designated offshore oil extraction area are as follows: Standardize the sea surface climate stability index, sea surface evaporation rate, wave height, and sea surface photosynthetic intensity at each time point within the designated offshore oil extraction area; and perform comprehensive analysis (i.e., weighted processing) on ​​the standardized sea surface climate stability index, sea surface evaporation rate, wave height, and sea surface photosynthetic intensity at each time point within the designated offshore oil extraction area to obtain the sea surface extraction stress index at each time point within the designated offshore oil extraction area.

[0035] The sea surface climate stability index measures the stability of the ocean surface environment. It is obtained by acquiring sea surface temperature (obtained through temperature sensors), sea surface air pressure (obtained through air pressure sensors), sea surface wind speed (obtained through anemometers), sea surface temperature reference value, sea surface air pressure reference value, and sea surface wind speed reference value. Difference analysis is then performed on each of these values ​​(e.g., the absolute value of the difference between the sea surface temperature value and the sea surface temperature reference value), and the results are standardized. The standardized results are then weighted and the reciprocal is taken. The final result is the sea surface climate stability index. The sea surface temperature reference value, sea surface air pressure reference value, and sea surface wind speed reference value can all be obtained from marine meteorological databases.

[0036] The sea surface evaporation rate is the amount of water vapor that evaporates from the seawater into the atmosphere per unit time in the offshore oil extraction area. A high evaporation rate means drastic weather changes, which can easily affect the stability of oil platforms and the operating conditions of equipment.

[0037] The wave height value is the average of multiple wave heights (the vertical distance between wave crests and troughs) within the offshore oil extraction area. Each wave height can be obtained using an altimeter in a wave buoy. High waves can affect the stability of the oil platform, increasing platform swaying or tilting, and thus affecting operational safety.

[0038] The photosynthetic intensity of marine organisms is the intensity of photosynthesis carried out by phytoplankton (mainly phytoplankton) in the seawater of the marine oil extraction area. It is expressed as photosynthetically active radiation (PAR) and can be obtained by photosynthetic radiation sensors. Petroleum and its derivatives (such as toxic chemicals such as benzene and toluene) can easily have a toxic effect on the photosynthesis of phytoplankton, that is, affect the efficiency of phytoplankton in obtaining light energy and reduce the intensity of photosynthesis.

[0039] In this implementation plan, standardized processing transforms diverse environmental data into unified measurement standards, and weighted analysis allows for the reasonable quantification of the impact of each indicator on offshore extraction operations. For example, the effects of wave height and sea surface evaporation rate on platform stability and operating conditions are effectively integrated, while changes in sea surface photosynthetic intensity can indicate potential environmental pollution risks, thus accurately predicting the impact of environmental changes on extraction safety. Secondly, comprehensive analysis of multiple environmental parameters helps provide a comprehensive emergency response framework for oil extraction under different environmental changes, thereby offering a more flexible and adaptable early warning mechanism and timely identification of potential operational risks. Finally, monitoring sea surface photosynthetic intensity can effectively assess the impact of environmental pollution (such as oil spills) on phytoplankton photosynthesis, thereby predicting potential threats to marine ecosystems from extraction activities. By monitoring these key environmental factors, events that may harm the environment can be detected and warned in advance, thus ensuring the safety and environmental protection of oil extraction activities.

[0040] Specifically, the time-series data of mid-ocean monitoring includes the mid-ocean density jump index, ocean current intensity, oxygen-sulfur corrosion index, and water shear stress. The specific steps to obtain the mid-ocean operational disturbance index for each time point within the designated offshore oil extraction area are as follows: Normalize the mid-ocean density jump index, ocean current intensity, oxygen-sulfur corrosion index, and water shear stress for each time point within the designated offshore oil extraction area (i.e., remove units); and comprehensively analyze the normalized mid-ocean density jump index, ocean current intensity, oxygen-sulfur corrosion index, and water shear stress for each time point within the designated offshore oil extraction area to obtain the mid-ocean operational disturbance index for each time point within the designated offshore oil extraction area.

[0041] The specific formula for calculating the mid-ocean operational disturbance index at each time point within a designated offshore oil extraction area is as follows: ;in, To define the first [area] within the offshore oil extraction zone Ocean mid-level operational disturbance index at a given time point To define the first [unit / area] within the offshore oil exploration area after normalization. The intermediate dense jump index at each time point is the dense jump adjustment coefficient stored in the database. To define the first [unit / area] within the normalized offshore oil exploration area Ocean current intensity values ​​at each time point, The adjustment coefficients for ocean current intensity stored in the database. To define the first [unit / area] within the normalized offshore oil exploration area Oxy-sulfur corrosion index at a given time point The oxygen-sulfur corrosion adjustment coefficient is stored in the database. To define the first [unit / area] within the normalized offshore oil exploration area The water shear stress value at each time point The water shear adjustment factor (which is not 0) is stored in the database. 1, 2, 3, ... , The number of time points.

[0042] It needs to be explained that, , , , The following steps can be taken to obtain the initial influence weights of each variable (mid-layer density jump index, ocean current intensity value, oxygen-sulfur corrosion index, and water shear stress value) on the mid-layer ocean disturbance index based on historical data and statistical regression analysis. Then, the range of coefficient values ​​is adjusted using sensitivity analysis to assess the stability and applicability of these parameters to the formula output. Next, the weights are further fitted through model optimization (such as machine learning algorithms) to ensure that the formula accurately reflects the actual state of mid-layer ocean disturbance. The coefficients are then fine-tuned based on the characteristics of different regions to ensure that they are applicable to specific mid-layer ocean disturbance assessment needs.

[0043] Among them, the ocean current intensity value is the flow intensity of seawater in the middle layer of the seawater in the designated offshore oil extraction area. Strong ocean currents can easily affect the stability of pipelines. It can be obtained by an acoustic Doppler current profiler, that is, the equipment will continuously emit sound waves and measure the water flow rate at different depths, and perform averaging to obtain the ocean current intensity.

[0044] The oxygen-sulfur corrosion index is used to assess the corrosive effects of dissolved oxygen and sulfide concentrations in the marine environment on pipelines and other equipment in the middle layer of seawater. It can be obtained by acquiring the dissolved oxygen concentration (which can be obtained through dissolved oxygen sensors) and hydrogen sulfide concentration (which can be obtained through chemical sensors) in the middle layer of seawater in a designated offshore oil extraction area, and then standardizing them. Based on the standardized results, a weighted average is performed, but the reciprocal of the standardized dissolved oxygen concentration value is taken during the weighting process, i.e., 1 / (1 + standardized dissolved oxygen concentration value). The result is the oxygen-sulfur corrosion index.

[0045] The water shear stress value is the shear force caused by the difference in flow velocity between water layers during the flow of seawater in the designated offshore oil extraction area. Under high-speed water flow, this may lead to increased friction, vibration, or movement of the pipeline, or even loss of pipeline stability. It can be obtained by acquiring the dynamic viscosity of the seawater mid-layer (i.e., the viscosity of the seawater mid-layer, which can be obtained based on the temperature and salinity values ​​of the mid-layer of the seawater using CTD sensors, weighted after standardization, and multiplied by the initial viscosity value of seawater stored in the data, such as 1.002 mPa·s at 20°C) and the water flow velocity at different depths (obtained by an acoustic Doppler current profiler), and performing difference analysis on the water flow velocity at different depths, applying the corresponding depth difference, and multiplying it by the dynamic viscosity. The result is the water shear stress value.

[0046] The following is a specific implementation example for calculating the mid-level operational disturbance index at each time point within a designated offshore oil extraction area. The available data includes the mid-level density jump index, ocean current intensity, oxygen-sulfur corrosion index, and water shear stress at five time points within the designated offshore oil extraction area, as shown in Table 1. Figures 2-5 As shown:

[0047] Table 1. Example of mid-ocean strata monitoring time series data at 5 time points within a designated offshore oil extraction area.

[0048]

[0049] Normalize the data examples in Table 1 to obtain Table 2:

[0050] Table 2. Examples of normalized mid-ocean strata monitoring time-series data for five time points within a designated offshore oil exploration area.

[0051]

[0052] The adjustment coefficients of the dense transition stored in the database Approximately 0.193;

[0053] The ocean current intensity adjustment factor stored in the database is approximately 0.267;

[0054] Oxygen-sulfur corrosion adjustment coefficients stored in the database Approximately 0.351;

[0055] Water shear adjustment coefficients stored in the database Approximately 0.684;

[0056] Substituting the data from Table 2 and the aforementioned adjustment coefficients into the specific formula for calculating the mid-ocean operational disturbance index at each time point within the designated offshore oil exploration area, we obtain:

[0057] The ocean mid-level operational disturbance index at time point 1 within the offshore oil exploration area is set to approximately 0.256.

[0058] The ocean mid-level operational disturbance index at time point 2 within the offshore oil exploration area is set to approximately 0.237.

[0059] The ocean mid-level operational disturbance index at time point 3 within the offshore oil exploration area is set to approximately 0.198.

[0060] The ocean mid-level operational disturbance index at time point 4 within the offshore oil exploration area is set to approximately 0.214.

[0061] The disturbance index of mid-ocean operations at time point 5 within the offshore oil exploration area is set to approximately 0.249.

[0062] The specific steps for obtaining the mid-layer density jump index at each time point within a designated offshore oil production area are as follows: Obtain temperature values ​​(which can be obtained using a CTD sensor) and salinity values ​​(which can be obtained using a CTD sensor) at several depths within the designated offshore oil production area at each time point, and perform comprehensive analysis (i.e., standard deviation processing) to obtain the temperature gradient value and salinity gradient value at each time point within the designated offshore oil production area; then perform comprehensive analysis (i.e., first perform standardization processing, and then perform weighted processing based on the standardization results) on the temperature gradient value and salinity gradient value at each time point within the designated offshore oil production area to obtain the mid-layer density jump index at each time point within the designated offshore oil production area.

[0063] This implementation plan normalizes and comprehensively analyzes multiple key environmental parameters to fully reflect the combined impact of the mid-ocean environment on oil extraction operations. This comprehensive analysis provides a more accurate disturbance index, making the early warning system more effective. Secondly, through regression analysis, sensitivity analysis, and model optimization, the influence weights of each parameter can be dynamically adjusted to ensure that the mid-ocean disturbance index accurately reflects real-time changes in the marine environment, thereby improving the adaptability and accuracy of monitoring and early warning. Adjusting coefficients ensures its stability and applicability under different environmental conditions. Real-time monitoring and analysis can provide early warnings of potential equipment problems during offshore oil extraction, allowing for proactive measures to reduce operational risks. Finally, through comprehensive analysis and optimized calculations, potential risk factors can be accurately assessed, providing a safer operating environment for oil extraction, reducing the likelihood of accidents, and ultimately improving the safety and long-term sustainability of oil extraction operations.

[0064] Specifically, such as Figure 6As shown, the time-series data for ocean bottom monitoring includes seafloor topographic image data at each time point, seafloor pulsation risk index (used to measure the stability of seafloor strata and the risk of energy release during oil extraction), and seafloor microbial activity index. Specifically, the seafloor topographic image data consists of the pixel value and two-dimensional coordinates of each pixel in the seafloor topographic map. The specific steps for obtaining the comprehensive seafloor extraction risk index at each time point within a designated marine oil extraction area are as follows: The seafloor topographic image data at each time point within the designated marine oil extraction area is input into a pre-trained topographic recognition model for predictive analysis to obtain the seafloor topographic extraction risk index at each time point within the designated marine oil extraction area; a comprehensive analysis is then performed on the seafloor topographic extraction risk index, seafloor disaster index, and seafloor microbial activity index at each time point within the designated marine oil extraction area to obtain the comprehensive seafloor extraction risk index at each time point within the designated marine oil extraction area.

[0065] The specific steps for obtaining the seabed pulsation risk index at each time point within a designated offshore oil extraction area are as follows: Obtain the seabed pressure gradient value, seabed vibration value, sedimentary reverse intensity value, and seabed vibration value at each time point within the designated offshore oil extraction area, and perform standardization processing; Perform comprehensive analysis (weighted processing) on ​​the standardized seabed pressure gradient value, seabed vibration value, sedimentary reverse intensity value, and seabed vibration value at each time point within the designated offshore oil extraction area to obtain the seabed pulsation risk index at each time point within the designated offshore oil extraction area.

[0066] The seabed pressure gradient value is the standard deviation of the pressure values ​​at different depths in the seabed within a set offshore oil and gas extraction area. Pressure values ​​at different depths can be obtained through pressure sensors. A high pressure gradient can easily lead to uneven pressure in seawater or oil and gas layers, thereby affecting the flow of oil and gas during drilling.

[0067] The seabed microbial activity index represents the activity level of the microbial community in the seabed environment within a designated offshore oil extraction area. It can easily lead to seabed corrosion, particularly of metal equipment, thus affecting the lifespan and stability of oil and gas extraction equipment. The seabed microbial activity index is obtained by acquiring methane, carbon dioxide, ammonia, and hydrogen chloride concentrations at multiple locations on the seabed within the designated offshore oil extraction area, standardizing these values, weighting the results based on the standardization, and then averaging them. The methane, carbon dioxide, ammonia, and hydrogen chloride concentrations can all be obtained using underwater gas sensors.

[0068] The seabed vibration value is the intensity of seabed vibration activity within the designated offshore oil and gas extraction area. It can easily affect the stability of equipment and oil and gas reservoirs and can be obtained through vibration sensors.

[0069] The backscattering intensity of sediments can be obtained by sonar detectors. After sound waves are emitted to the seabed, some of them are reflected back to the detector by seabed sediments. The intensity of the reflected sound waves (i.e., backscattering intensity) is used to reflect the density and fluidity of sediments. A lower sediment backscattering intensity value indicates that the sediments are loose and there is a potential risk of landslides or collapses, which poses a threat to the safety of seabed oil and gas extraction operations and the stability of equipment.

[0070] The specific steps for calculating the comprehensive seabed extraction risk index at each time point within a defined offshore oil extraction area are as follows: ;in, To define the first [area] within the offshore oil extraction zone The comprehensive risk index of seabed mining at a given time point. To define the first [area] within the offshore oil extraction zone Risk index of seabed topography mining at a given time point. The adjustment coefficients for the seabed topography stored in the database. To define the first [area] within the offshore oil extraction zone The risk index of seabed pulsation at a specific time point. This refers to the pulsation risk adjustment coefficient stored in the database. To define the first [area] within the offshore oil extraction zone Seabed microbial activity index at a specific time point This refers to the adjustment coefficient for microbial activity stored in the database. These are the collaborative adjustment coefficients stored in the database (used to adjust the combined effects of the seabed topography adjustment coefficient and the seabed pulsation risk index). 1, 2, 3, ... , The number of time points.

[0071] It needs to be explained that in the formula This item is used to adjust the interaction between the seabed topography adjustment coefficient and the seabed pulsation risk index, so as to avoid the overall risk index of seabed mining being too high or too low.

[0072] The expression for the Tanh function is: ,in, It is a natural constant, and in this embodiment it is taken as 2.71, with a domain of (-∞, +∞) and a range of (-1, 1).

[0073] , , , The following steps can be taken to obtain the following: Based on historical data, determine the initial influence weights of each variable (seafloor topography mining risk index, seafloor disaster index, and seafloor microbial activity index) on the comprehensive seafloor mining risk index through statistical regression analysis. Then, use sensitivity analysis to adjust the range of coefficient values ​​to assess the stability and applicability of these parameters to the formula output. Next, further fit the weights through model optimization (such as machine learning algorithms or multi-objective optimization) to ensure that the formula can accurately reflect the actual state of comprehensive seafloor mining risk. Fine-tune the coefficients based on the characteristics of different regions to ensure that they are applicable to specific comprehensive seafloor mining risk assessment needs.

[0074] This implementation plan combines multiple factors such as seabed topography, seabed pulsation risk, and microbial activity to conduct seabed risk analysis from multiple perspectives. This helps to comprehensively understand the complexity of the seabed environment, thereby ensuring the accurate identification and prediction of potential risks. Secondly, by analyzing the seabed topography mining risk index at each time point, the accuracy of risk assessment is improved. Real-time acquisition of data such as seabed pressure gradient, vibration values, and sediment backscattering intensity, along with standardization and weighted analysis, allows for real-time monitoring of changes in the seabed environment, rapid detection of potential risks, and prevention of accidents. Finally, by introducing adjustment coefficients and co-adjustment coefficients, the weight of each risk index can be flexibly adjusted according to different seabed environmental characteristics and actual conditions, ensuring the stability and adaptability of the calculation results. This enhances the flexibility and reliability of monitoring and early warning, provides data support for equipment maintenance and protection, extends equipment lifespan, and ensures the stable operation of oil extraction.

[0075] Specifically, the terrain recognition model uses a convolutional neural network (CNN). The CNN includes an input layer, several convolutional layers, activation layers, pooling and flattening layers, and a risk output layer. The specific steps for obtaining the seabed terrain risk index for each time point within a designated offshore oil extraction area are as follows: In the input layer of the CNN, seabed terrain image data (i.e., the pixel value and two-dimensional coordinates of each pixel in the seabed terrain map) for each time point within the designated offshore oil extraction area are received and preprocessed. In the convolutional layers of the CNN, feature extraction processing is performed on the preprocessed seabed terrain image data for each time point within the designated offshore oil extraction area. (That is, by performing a convolution operation between a convolution kernel and the input image, local features are extracted from the image, such as seabed cracks, faults, trenches, submarine ridges, etc.), to obtain the seabed topographic feature map at each time point within a designated offshore oil extraction area; in the activation layer of the convolutional neural network, a nonlinear transformation is performed on the seabed topographic feature map at each time point within the designated offshore oil extraction area (using the ReLU activation function to introduce nonlinearity, enabling the network to learn complex topographic patterns and enhance its sensitivity to important features such as extraction risks); in the pooling and flattening layer of the convolutional neural network, the activated designation of the designated offshore oil extraction area is further processed. The seafloor topography feature map at each time point within the domain is subjected to pooling and flattening processing (using max pooling to reduce the spatial size of the feature map while preserving the most important feature information; that is, by sliding a fixed-size window on the feature map and merging the elements within the window into a single value, i.e., the maximum value within the pooling window is retained, thus obtaining a dimensionality-reduced feature map, which is then converted into a one-dimensional feature vector). This yields the seafloor topography feature vector at each time point within the designated offshore oil exploration area (including but not limited to the degree of trench erosion, the complexity of the seafloor topography, and the stability of faults). In the risk output layer of the convolutional neural network, the risk of the designated offshore oil exploration area is then processed. The seabed topographic feature vectors at each time point within the region are fused (different topographic features in the seabed topographic feature vectors are weighted, and a nonlinear activation function is applied to the weighted features so that the network can capture the complex nonlinear relationships between topographic features. Then, a fully connected layer is passed to map the input features to a high-dimensional space, thereby further capturing the relationships between features. After the fully connected layer, a regression layer is passed to map it to a continuous numerical output, namely the seabed topographic mining risk index, with a value range between 0 and 1), to obtain the seabed topographic mining risk index at each time point within the designated offshore oil extraction area.

[0076] The input layer is used to accept raw seabed topographic image data as input to the neural network.

[0077] Convolutional layers are used to extract features from input image data. They find local features in the image through convolution operations and generate seabed topographic feature maps.

[0078] Activation layers are used to perform nonlinear transformations on the features extracted by the convolutional layers (i.e., seabed topographic feature maps), enabling the model to learn more complex patterns.

[0079] Pooling flattening layers are used to perform pooling operations (such as max pooling or average pooling) to reduce the spatial size of the seabed topographic feature map (downsampling), and then flatten the pooled feature map into a one-dimensional vector of seabed topographic features.

[0080] The risk output layer is used to generate the final seabed topography mining risk index based on a one-dimensional vector of seabed topographic features.

[0081] Furthermore, the pre-training process of a convolutional neural network is as follows:

[0082] Obtain a dataset of seabed topographic images, including several seabed topographic images (e.g., seabed cracks, faults, gullies, etc.) and corresponding annotation information (e.g., hazard level, risk index, etc.). Divide the seabed topographic image dataset into a training set and a validation set, usually in a ratio of 80%:20%.

[0083] The convolutional neural network is initialized by initializing the weights and biases in the network to ensure that the network can effectively learn features. Initialization methods include Xavier initialization and He initialization, such as initializing the weights of the convolutional layers (using He initialization), and initializing the bias terms to zero. Then, the ReLU activation function is used to increase the nonlinearity of the network.

[0084] Training is performed on a training set, with a set number of training iterations (e.g., 50 or 100 iterations). In each training iteration, forward propagation (the network acquires a batch of data from the training set, for example, inputting one seabed topographic image at a time and passing it through convolutional layers, activation layers, pooling layers, etc., to generate a seabed topographic mining risk index), loss calculation (comparing the network's predicted values, i.e., the predicted seabed topographic mining risk index, with the true label value, to calculate the model's loss; commonly used loss functions include mean squared error, cross-entropy loss, etc.), and backpropagation (calculating the gradient in the network based on the loss function and updating the model parameters through the optimizer).

[0085] After each training cycle, an evaluation analysis is performed based on the validation set. Forward propagation is performed using images from the validation set to calculate the loss value between the predicted seabed topography mining risk index and the actual seabed topography mining risk index. The loss function value and accuracy of the model on the validation set are also calculated to evaluate the model's performance.

[0086] Adjusting model parameters based on evaluation results: If the loss on the validation set does not decrease or the accuracy does not improve, it may be necessary to adjust the network's hyperparameters (such as learning rate, network structure, etc.). If the performance on the validation set is poor, hyperparameters such as learning rate, batch size, and number of convolutional layers can be adjusted to improve training results. If the loss on the validation set does not improve significantly within several training cycles, training should be stopped to prevent overfitting. Based on the training and validation results, the structure of the model, such as the number of layers and the size of the convolutional kernels, can be adjusted.

[0087] The training process ends when the training is complete and the loss and accuracy on the validation set meet the expected standards, resulting in a well-trained network model.

[0088] In this implementation scheme, the convolutional neural network (CNN) can automatically extract important features from seabed topographic images, such as cracks, faults, and gullies, thereby obtaining detailed topographic features and improving the accuracy of risk assessment. Furthermore, through nonlinear transformations in the activation layer, the CNN can capture complex nonlinear relationships in the seabed topographic data. The use of nonlinear activation functions enhances the model's sensitivity to complex mining risks. The pooling and flattening layer reduces the spatial dimension of seabed topographic image features through dimensionality reduction while retaining important information and avoiding overfitting. Secondly, through the risk output layer of the CNN, the model can comprehensively process the extracted features and output a seabed topographic mining risk index, which helps assess the mining risk at each time point and provides a scientific basis for decision-makers. Finally, the training process of the CNN can be optimized based on actual data, and the model parameters for mining risk assessment can be optimized through learning from a large amount of data, thereby dynamically adjusting risk predictions to adapt to the complex and changing seabed environment, ultimately enabling the model to provide more accurate seabed mining risk predictions.

[0089] Specifically, the steps for intelligent oil extraction early warning based on the oil extraction early warning index at each time point within a designated offshore oil extraction area are as follows: Trend analysis is performed on the oil extraction early warning index at each time point within the designated offshore oil extraction area to obtain several sets of oil extraction early warning index change rates within the designated offshore oil extraction area; a comprehensive analysis is then performed on the change rates of each set of oil extraction early warning indices within the designated offshore oil extraction area and the oil extraction early warning index at each time point (i.e., the oil extraction early warning index at each time point is averaged to obtain the mean oil extraction early warning index, and the change rates of each set of oil extraction early warning indices are weighted, and the weighted result is summed with the mean oil extraction early warning index), to obtain the predicted oil extraction early warning index within the designated offshore oil extraction area, which is considered the oil extraction early warning index for the next time point within the designated offshore oil extraction area; based on the oil extraction early warning index for the next time point within the designated offshore oil extraction area, preset early warning measures are taken, the specific steps of which are as follows: The oil extraction early warning index for the next time point within the designated offshore oil extraction area is compared with a preset oil extraction early warning index threshold set, the oil extraction early warning index threshold set including the first oil extraction... The system comprises three thresholds: a first oil extraction warning index threshold (to measure whether an extraction risk exists), a second oil extraction warning index threshold (to measure whether the extraction risk is low or medium), and a third oil extraction warning index threshold (to measure whether a high risk exists). If the oil extraction warning index at the next time point in the offshore oil extraction area is lower than the first oil extraction warning index threshold, it is marked as safe, and monitoring continues. If the oil extraction warning index at the next time point in the offshore oil extraction area is higher than or equal to the first oil extraction warning index threshold but lower than the second oil extraction warning index threshold, it is marked as low risk, a low-risk warning notice is issued to relevant personnel, and the preset first warning measures are implemented (i.e., increasing the frequency of data monitoring and performing preventative maintenance on key equipment to avoid higher risks due to equipment aging or problems). If the oil extraction warning index at the next time point in the offshore oil extraction area is higher than or equal to the second oil extraction warning index threshold but lower than the third oil extraction warning index threshold, it is marked as medium risk, a medium-risk warning notice is issued to relevant personnel, and the preset second warning measures are implemented (i.e., reducing the extraction speed and activating underwater robots to conduct a more comprehensive inspection of key equipment to eliminate possible malfunctions).If the oil exploration early warning index at the next set time point in the offshore oil exploration area exceeds the third oil exploration early warning index threshold, it will be marked as high risk. A high-risk early warning notice will be issued to relevant personnel, and the pre-set third early warning measures will be taken (i.e., immediate production stoppage or suspension of operations to ensure the safety of personnel and equipment, activation of a comprehensive emergency plan, including evacuation, environmental protection, accident handling, and other emergency response measures, followed by reporting the situation to the superior management department and the offshore oil safety regulatory agency, and subsequent handling as required).

[0090] This implementation plan utilizes trend analysis of the oil extraction early warning index to identify potential risks in advance, enabling preventative measures to be taken before dangerous situations occur. This reduces the probability of accidents and protects the safety of extraction equipment, personnel, and the environment. By setting different risk levels and early warning thresholds (e.g., low, medium, and high risk), different levels of response measures can be taken based on the severity of the risk, avoiding overreaction or underreaction. Secondly, by predicting the oil extraction early warning index at the next time point, production strategies can be adjusted in real time. For example, when the early warning index enters the low-risk zone, data monitoring and equipment maintenance can be strengthened; when it reaches the high-risk zone, production can be immediately halted and emergency response measures can be initiated. Finally, by establishing emergency response mechanisms, such as evacuation, environmental protection, and accident handling, rapid response can be achieved in the event of emergencies, reducing damage to the ecological environment, minimizing the need for human intervention, and improving management efficiency and accuracy.

[0091] See also Figure 7 This invention provides a technical solution: a marine three-dimensional stratified rights-based project monitoring and early warning system, comprising: a marine monitoring time series acquisition module, used to continuously acquire marine monitoring time series datasets within a designated marine oil extraction area, including marine surface monitoring time series data, marine mid-layer monitoring time series data, and marine bottom monitoring time series data; a marine stratification feature analysis module, used to perform feature analysis on the marine monitoring time series datasets within the designated marine oil extraction area to obtain a monitoring and early warning index set for each time point within the designated marine oil extraction area, including a sea surface extraction stress index, a marine mid-layer operation disturbance index, and a seabed extraction comprehensive risk index; a comprehensive early warning assessment module, used to comprehensively analyze the monitoring and early warning index set for each time point within the designated marine oil extraction area to obtain an oil extraction early warning index for each time point within the designated marine oil extraction area; and an intelligent early warning module, used to provide intelligent oil extraction early warning based on the oil extraction early warning index for each time point within the designated marine oil extraction area.

[0092] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0093] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for monitoring and early warning of marine three-dimensional stratified weighted projects, characterized in that, Includes the following steps: Continuously acquire marine monitoring time-series datasets within a designated offshore oil extraction area, including marine surface monitoring time-series data, mid-ocean strata monitoring time-series data, and bottom-ocean monitoring time-series data; Feature analysis was performed on the marine monitoring time series datasets within the designated offshore oil extraction area to obtain a set of monitoring and early warning indices for each time point within the designated offshore oil extraction area, including the sea surface extraction stress index, the mid-ocean strata operation disturbance index, and the seabed extraction comprehensive risk index. The ocean bottom monitoring time-series data includes seabed topographic image data, seabed pulsation risk index, and seabed microbial activity index at each time point. Specifically, the seabed topographic image data consists of the pixel value and two-dimensional coordinates of each pixel in the seabed topographic map. The specific steps for obtaining the comprehensive seabed extraction risk index at each time point within a designated offshore oil extraction area are as follows: The seabed topographic image data at each time point within the designated offshore oil extraction area are input into a pre-trained topographic recognition model for predictive analysis, thereby obtaining the seabed topographic extraction risk index at each time point within the designated offshore oil extraction area. A comprehensive analysis of the seabed topography risk index, seabed disaster index, and seabed microbial activity index at each time point within a designated offshore oil extraction area is conducted to obtain the comprehensive seabed extraction risk index at each time point within the designated offshore oil extraction area. The terrain recognition model specifically uses a convolutional neural network, which includes an input layer, several convolutional layers, an activation layer, a pooling and flattening layer, and a risk output layer. The specific steps for obtaining the seabed terrain mining risk index at each time point within a designated offshore oil extraction area are as follows: In the input layer of the convolutional neural network, seabed topographic image data at each time point within a designated offshore oil extraction area are received and preprocessed. In the convolutional layer of the convolutional neural network, feature extraction processing is performed on the seabed topographic image data at each time point within the preprocessed marine oil extraction area to obtain the seabed topographic feature map at each time point within the designated marine oil extraction area. In the activation layer of the convolutional neural network, a nonlinear transformation is performed on the seabed topographic feature map at each time point within a defined offshore oil extraction area. In the pooling and flattening layer of the convolutional neural network, the seabed topographic feature map at each time point in the designated offshore oil extraction area after activation is pooled and flattened to obtain the seabed topographic feature vector at each time point in the designated offshore oil extraction area. In the risk output layer of the convolutional neural network, the seabed topographic feature vectors at each time point within the designated offshore oil extraction area are fused to obtain the seabed topographic extraction risk index at each time point within the designated offshore oil extraction area. A comprehensive analysis of the monitoring and early warning index set for each time point within a designated offshore oil extraction area yields the oil extraction early warning index for each time point within the designated offshore oil extraction area. Intelligent oil extraction early warning is based on an oil extraction early warning index set for each time point in the offshore oil extraction area. The specific formula for calculating the oil extraction early warning index at each time point within a designated offshore oil extraction area is as follows: ； in, , , , The order is as follows: within the designated offshore oil exploration area... Oil extraction early warning index, sea surface extraction stress index, mid-ocean strata disturbance index, and seabed extraction comprehensive risk index at various time points. To define the first [area] within the offshore oil extraction zone The sea surface mining stress index at a given time point , , , , , , The following are, in order, the sea surface stress coefficient, sea surface stress adjustment coefficient, mid-level disturbance coefficient, mid-level disturbance adjustment coefficient, seabed risk coefficient, seabed risk adjustment coefficient, and interaction coefficient stored in the database. The smoothing coefficients are stored in the database. , 1, 2, 3, ... , The number of time points.

2. The marine three-dimensional stratified rights setting project monitoring and early warning method according to claim 1, characterized in that, The time-series data of ocean surface monitoring includes the sea surface climate stability index, sea surface evaporation rate, wave height, and sea surface photosynthetic intensity at each time point. The specific steps to obtain the sea surface extraction stress index at each time point within a designated offshore oil extraction area are as follows: Standardize the sea surface climate stability index, sea surface evaporation rate, wave height, and sea surface photosynthetic intensity at each time point within the designated offshore oil extraction area; The sea surface climate stability index, sea surface evaporation rate, wave height, and sea surface photosynthetic intensity of marine organisms at each time point within the designated offshore oil extraction area were comprehensively analyzed to obtain the sea surface extraction stress index at each time point within the designated offshore oil extraction area.

3. The marine three-dimensional stratified rights setting project monitoring and early warning method according to claim 1, characterized in that, The time-series data for mid-ocean monitoring includes mid-ocean density jump index, ocean current intensity value, oxygen-sulfur corrosion index, and water shear stress value. The specific steps for obtaining the mid-ocean operational disturbance index at each time point within the designated offshore oil extraction area are as follows: The mid-layer density jump index, ocean current intensity, oxygen-sulfur corrosion index, and water shear stress value at each time point within the designated offshore oil extraction area were normalized. The mid-layer density jump index, ocean current intensity, oxygen-sulfur corrosion index, and water shear stress value at each time point within the normalized offshore oil extraction area were comprehensively analyzed to obtain the mid-layer operational disturbance index at each time point within the designated offshore oil extraction area.

4. The marine three-dimensional stratified rights setting project monitoring and early warning method according to claim 3, characterized in that, The specific steps for obtaining the mid-layer compaction index at each time point within a designated offshore oil extraction area are as follows: The temperature and salinity values ​​at several depths at each time point within a designated offshore oil extraction area are obtained and comprehensively analyzed to obtain the temperature gradient and salinity gradient values ​​at each time point within the designated offshore oil extraction area. The temperature gradient and salinity gradient values ​​at each time point within the designated offshore oil extraction area were comprehensively analyzed to obtain the mid-layer compaction index at each time point within the designated offshore oil extraction area.

5. The marine three-dimensional stratified rights setting project monitoring and early warning method according to claim 1, characterized in that, The specific steps for calculating the comprehensive seabed extraction risk index at each time point within a defined offshore oil extraction area are as follows: ； in, , , , The order is as follows: within the designated offshore oil exploration area... The risk indices for seabed mining at various time points include: comprehensive risk index for seabed topography mining, risk index for seabed pulsation, and risk index for seabed microbial activity. To define the first [area] within the offshore oil extraction zone Risk index of seabed topography mining at a given time point. , , , The following are, in order, the seabed topography adjustment coefficient, the pulsation risk adjustment coefficient, the microbial activity adjustment coefficient, and the synergistic adjustment coefficient stored in the database. 1, 2, 3, ... , The number of time points.

6. The marine three-dimensional stratified rights setting project monitoring and early warning method according to claim 1, characterized in that, The specific steps for intelligent oil extraction early warning based on the oil extraction early warning index set for each time point within the offshore oil extraction area are as follows: Trend analysis was performed on the oil extraction early warning index at each time point within the designated offshore oil extraction area to obtain the change rate of several sets of oil extraction early warning indices within the designated offshore oil extraction area. The change rate of each set of oil extraction early warning indices and the oil extraction early warning index at each time point in the set offshore oil extraction area are comprehensively analyzed to obtain the predicted oil extraction early warning index in the set offshore oil extraction area, which is regarded as the oil extraction early warning index for the next time in the set offshore oil extraction area. Based on the oil extraction early warning index set for the next time point in the offshore oil extraction area, corresponding pre-set early warning measures are taken.

7. A marine three-dimensional stratified weighting project monitoring and early warning system, employing the marine three-dimensional stratified weighting project monitoring and early warning method according to any one of claims 1-6, characterized in that, include: The marine monitoring time series acquisition module is used to continuously acquire marine monitoring time series datasets within a designated marine oil extraction area, including marine surface monitoring time series data, mid-ocean stratum monitoring time series data, and bottom-ocean monitoring time series data. The marine stratification feature analysis module is used to perform feature analysis on the marine monitoring time series datasets within a designated marine oil extraction area, and obtain the monitoring and early warning index set for each time point within the designated marine oil extraction area, including the sea surface extraction stress index, the mid-ocean operation disturbance index, and the seabed extraction comprehensive risk index. The comprehensive early warning and assessment module is used to comprehensively analyze the monitoring and early warning index set for each time point in the designated offshore oil extraction area, and obtain the oil extraction early warning index for each time point in the designated offshore oil extraction area. The intelligent early warning module is used to provide intelligent oil extraction early warning based on the oil extraction early warning index at each time point within a set offshore oil extraction area.

Images