A system and method for underwater three-dimensional topographic measurement of oil fields
Through the collaborative work of the regional division and acquisition equipment optimization module, combined with multi-source data fusion technology, the problems of complex terrain and low data acquisition efficiency of underwater oil fields are solved, and the accuracy and timeliness of the underwater three-dimensional terrain simulation model of oil fields are achieved. Ability to quickly identify and update abnormal areas.
Patent Information
- Application Number
- CN202411116852.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-15
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-08-15
AI Technical Summary
The underwater oil field has complex terrain and low data collection efficiency, and it is impossible to quickly identify and screen out abnormal areas, resulting in insufficient accuracy and timeliness of the underwater three-dimensional terrain simulation model of the oil field.
Through the coordinated work of the area division module and the acquisition equipment optimization module, multi-source data fusion technology is used to scan and update abnormal areas in a targeted manner to build and optimize the underwater three-dimensional terrain simulation model of oil fields. Specific steps include turbidity monitoring, equipment parameter adjustment, data acquisition, data preprocessing, three-dimensional modeling, abnormal area screening and multi-source data fusion.
It improves the accuracy and efficiency of data acquisition, significantly enhances the reliability and real-timeness of the underwater three-dimensional terrain simulation model of the oil field, and can quickly identify and screen out abnormal areas, ensuring the accuracy and timeliness of the model.
Smart Images

Figure CN118884459B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of topographic surveying, and in particular to an underwater three-dimensional topographic surveying system and method for an oil field. Background Art
[0002] By accurately measuring the underwater three-dimensional topography of oil fields, detailed topographic data support can be provided for oil field development. These data are helpful in formulating scientific and reasonable development plans, optimizing oil well layout, reducing uncertainty in the drilling process, and thus improving the extraction efficiency of oil fields. At the same time, understanding the complexity and potential risks of underwater topography can help prevent drilling accidents and environmental pollution, and ensure the safety of oil field development.
[0003] In marine projects such as submarine pipeline laying and submarine cable maintenance, underwater 3D topographic measurement systems can provide key topographic data support. These data help engineers accurately assess the difficulty and risk of the project, design reasonable construction plans, and ensure the safety and reliability of the project. In addition, by monitoring underwater topographic changes in real time, potential safety hazards can be discovered and dealt with in a timely manner to ensure the smooth progress of marine projects.
[0004] However, in actual use, it still has many shortcomings, such as the complex terrain of underwater oil fields, low data collection efficiency, and the inability to quickly identify and screen out abnormal areas. Summary of the invention
[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides an oilfield underwater three-dimensional terrain measurement system and method, which solves the problems of complex underwater oilfield terrain and low data acquisition efficiency through the coordinated work of a region division module and an acquisition equipment optimization module; utilizes multi-source data fusion technology to perform targeted scanning and updating of abnormal areas, thereby ensuring the accuracy and timeliness of the oilfield underwater three-dimensional terrain simulation model, and solving the problem in the prior art that it is not possible to quickly identify and screen out abnormal areas and update them.
[0006] To achieve the above object, the present invention provides the following technical solution: an oilfield underwater three-dimensional topographic measurement system, comprising:
[0007] The area division module is used to divide the target underwater oil field into several areas and number them; suppose it is divided into n areas, and i represents the area number;
[0008] The data acquisition equipment optimization module identifies the turbidity of each area and outputs the equipment adjustment parameters of the data acquisition equipment based on the relationship between the turbidity and the preset value;
[0009] The underwater oilfield data acquisition module, after adjusting the parameters of the equipment, conducts several rounds of terrain scanning on the underwater oilfield to obtain several sets of three-dimensional information for each area; underwater cameras, sonar, and lidar equipment are used to collect data on the underwater environment; the collected three-dimensional information includes water depth, terrain, and underwater structure;
[0010] The data preprocessing module is used to clean, denoise and correct several groups of 3D information of each underwater area of the oil field, and use image processing and computer vision technologies such as optical flow and stereo matching to extract the 3D features of the underwater scene of each area;
[0011] The 3D modeling module constructs a 3D underwater terrain simulation model of the oil field based on the 3D features of the underwater scene in each area through rotation, scaling, translation and registration operations;
[0012] The 3D model abnormal area screening module is used to verify the similarity between the 3D reconstructed feature vector parameters and the actual feature vector parameters of the underwater 3D terrain simulation model of the oil field, and screen out abnormal areas;
[0013] The 3D model optimization and updating module scans the abnormal area to obtain multi-source data, and replaces the abnormal area of the oilfield underwater 3D terrain simulation model based on the fusion of multi-source data to obtain an updated oilfield underwater 3D terrain simulation model.
[0014] Preferably, when the turbidity of the i-th area exceeds a preset value, the device adjustment parameter of the collection device in the i-th area is output; the device adjustment parameter is obtained in the following manner:
[0015] Step S11, turbidity monitoring and determination: using a turbidity meter or a sensor integrated in a laser radar system to monitor the turbidity of the water body in the i-th area in real time; threshold comparison: comparing the real-time turbidity data with a preset turbidity threshold, if the turbidity exceeds the preset value, proceed to the next step;
[0016] Step S12, evaluating the impact of turbidity exceeding a preset value on three-dimensional accuracy, such as signal attenuation and reduction in signal-to-noise ratio; and determining target parameters that need to be adjusted, such as laser power and scanning wavelength, based on the evaluation results;
[0017] Step S13, laser power adjustment: according to the degree of turbidity exceeding, calculate or select the appropriate increase in laser output power; the constraint condition is that the increase in laser power does not exceed the safety limit and the hardware tolerance; according to the penetration ability and scattering characteristics of lasers of different wavelengths in turbid water bodies, select the appropriate wavelength combination and configure the multi-wavelength laser scanning system to obtain clearer images and more accurate distance measurement information;
[0018] Step S14, output device adjustment parameters: integrate the calculated laser output power increase and multi-wavelength laser scanning configuration; transmit the integrated adjustment parameters to the collection device in the i-th area through the control system; after receiving the adjustment parameters, the collection device adjusts its working state to adapt to the current turbidity level;
[0019] Step S15, monitoring and feedback: After the adjustment parameters are applied, the performance changes of the lidar system are continuously monitored to evaluate the impact of the adjustment parameters on the detection effect in turbid water bodies; if the adjustment effect is not ideal, the parameters are readjusted according to the monitoring results and evaluation feedback until a satisfactory detection effect is achieved.
[0020] Preferably, the laser output power increase is obtained by:
[0021] The laser output power under standard conditions is obtained, which is recorded as Pba; the laser output power that needs to be increased is Pinc; the total laser output power after adjustment is Pto;
[0022] By formula The signal attenuation factor A caused by turbidity is calculated. The signal attenuation factor A is dimensionless and greater than 1.
[0023] By formula The calculated laser output power that needs to be increased is Pinc;
[0024] By formula The adjusted total laser output power is calculated.
[0025] Preferably, the abnormal area is obtained by:
[0026] Based on manual measurement, actual measurement data of each area is obtained, and actual feature vector parameters are obtained from the actual measurement data; three-dimensional reconstruction feature vector parameters corresponding to the actual measurement data are obtained from the underwater three-dimensional terrain simulation model of the oil field; j represents the number of the feature vector parameter, and the number of the feature vector parameters is m;
[0027] Based on the actual feature vector parameters and the 3D reconstructed feature vector parameters, the similarity parameters of the feature vector parameters of each region are calculated. The feature similarity parameters of the i-th region are calculated as follows: , where sj represents the jth actual eigenvector parameter, and gj represents the jth 3D reconstructed eigenvector parameter;
[0028] By formula The accuracy index Zp of the underwater three-dimensional terrain simulation model of the oil field is calculated, where c i represents the feature similarity parameter of the i-th region, kc i represents the regional weight coefficient of the ith region. The calculation of the regional weight coefficient satisfies , η1 represents the accuracy influencing factor of manual measurement, η2 represents the preset 3D reconstruction accuracy influencing factor, and 0.95<η1<1.0, 0.85<η2<1.0, which is set based on actual conditions;
[0029] The accuracy index Zp of the oilfield underwater 3D terrain simulation model is compared with the preset value. When the accuracy index Zp is lower than the preset value, it indicates that the quality of the oilfield underwater 3D terrain simulation model is lower than the preset value C. 预 , re-collect data and build an underwater three-dimensional terrain simulation model of the oil field; when the accuracy index Zp is not lower than the preset value C 预 , indicating that the quality of the underwater three-dimensional terrain simulation model of the oilfield meets the preset requirements; the characteristic similarity parameters of each area are compared with the preset similarity one by one, and the abnormal areas are screened out; the areas with characteristic similarity parameters lower than the preset similarity are recorded as abnormal areas.
[0030] Preferably, the number of characteristic vector parameters is set based on the depth and geological instability of the region, and the number of characteristic vector parameters is obtained as follows:
[0031] Assume that the number of characteristic vector parameters required per unit volume in a stable oil field underwater is △C; through the formula The number of characteristic vector parameters is calculated; wherein m represents the number of characteristic vector parameters, V represents the regional volume, QF" represents the geological instability parameter after linear normalization, and SD represents the maximum water depth of the region; the geological instability parameter represents the deformation speed of the underwater topography of the oil field in the region.
[0032] Preferably, the three-dimensional model optimization and updating module comprises the following steps:
[0033] Step S21, collecting multi-source data: using high-precision scanning sonar and multi-beam sonar equipment to comprehensively scan the underwater abnormal area of the oil field, the acquired data includes the depth, slope, and morphological three-dimensional parameters of the seabed terrain, forming a first set of three-dimensional terrain data; using an underwater laser radar system to measure the seabed point cloud data by laser pulse reflection; the laser radar data provides a high-density three-dimensional point cloud to form a second set of three-dimensional terrain data; environmental sensor data: using temperature sensors, salinity meters, and flow rate meters deployed underwater to collect water temperature, salinity, and flow rate environmental parameters in real time;
[0034] Step S22, data preprocessing: using timestamps and GPS data to synchronize data from different sensors and time points; using calibration algorithms to convert data from different sources into a unified geographic coordinate system and projection system to achieve spatial calibration; analyzing the error range and stability of each data source under different conditions;
[0035] Step S23, according to the water depth, water flow velocity, sensor accuracy parameters, combined with historical data and theoretical models, fuzzy logic and neural network methods are used for comprehensive evaluation to calculate the positioning accuracy index of each data source; based on the positioning accuracy index, the weighted average method, entropy weight method or hierarchical analysis method are used to calculate the weight coefficient of each data source in different areas and depths;
[0036] Step S24, data fusion and model update: using the calculated weight coefficient, adopting weighted average or Bayesian fusion to fuse the sonar data, lidar data and other sensor data; using the fused three-dimensional reconstruction feature vector parameters to replace the abnormal area data of the oilfield underwater three-dimensional terrain simulation model, to obtain an updated oilfield underwater three-dimensional terrain simulation model;
[0037] Step S25, verifying the accuracy and reliability of the updated oilfield underwater three-dimensional terrain simulation model by comparing it with actual exploration data or other independent data sources.
[0038] Preferably, the positioning accuracy index of the sonar detection data is recorded as SZa, and the positioning accuracy index of the laser radar data is recorded as SZb;
[0039] The three-dimensional reconstruction feature vector parameters of sonar detection data are recorded as SZa_j, and the three-dimensional reconstruction feature vector parameters of lidar data are recorded as SZb_j;
[0040] By formula The calculated 3D reconstruction feature vector parameter is recorded as the weight coefficient of SZa_j; The calculated three-dimensional reconstruction feature vector parameters of the lidar data are recorded as the weight coefficient of SZb_j;
[0041] Based on the weight coefficient, through the formula The fused 3D reconstruction feature vector parameter R_s is calculated.
[0042] Preferably, the positioning accuracy index is obtained in the following manner:
[0043] A convolutional neural network model is constructed, and the water depth, water flow velocity, and sensor accuracy parameters are used as input features. The known positioning accuracy in historical data is used as the label, i.e., the target output. By training the convolutional neural network model, the nonlinear mapping relationship from the input parameters to the positioning accuracy index is learned. After the training is completed, the positioning accuracy index is obtained from the trained neural network model.
[0044] To achieve the purpose of the present invention, a method for measuring underwater three-dimensional topography in an oil field is provided, comprising the following steps:
[0045] Step 1: Divide the target underwater oil field into several areas and number them; divide them into n areas, and use i to represent the area number;
[0046] Step 2: Identify the turbidity of each area, and output the device adjustment parameters of the data acquisition device based on the relationship between the turbidity and the preset value;
[0047] Step 3: After adjusting the parameters of the equipment, several rounds of terrain scanning are carried out on the underwater oil field to obtain several sets of three-dimensional information of each area; underwater cameras, sonar, and lidar equipment are used to collect data on the underwater environment; the collected three-dimensional information includes water depth, terrain, and underwater structure;
[0048] Step 4: Clean, denoise and correct several groups of 3D information of each underwater area of the oil field, and use image processing and computer vision techniques, such as optical flow method and stereo matching, to extract the 3D features of the underwater scene of each area;
[0049] Step 5: construct an underwater 3D terrain simulation model of the oil field based on the 3D features of the underwater scene in each area through rotation, scaling, translation and registration operations;
[0050] Step 6: verify the similarity between the three-dimensional reconstructed feature vector parameters and the actual feature vector parameters of the underwater three-dimensional terrain simulation model of the oil field, and screen out abnormal areas;
[0051] Step 7: Scan the abnormal area to obtain multi-source data, and replace the abnormal area of the oilfield underwater three-dimensional terrain simulation model based on the fusion of the multi-source data to obtain an updated oilfield underwater three-dimensional terrain simulation model.
[0052] The beneficial effects of the present invention are:
[0053] The oilfield underwater three-dimensional topographic measurement system provided by the present invention solves the problems of complex underwater oilfield topography and low data acquisition efficiency through the coordinated work of the regional division module and the acquisition equipment optimization module; firstly, according to the actual situation of the oilfield, it is intelligently divided into multiple regions, and the parameters of the data acquisition equipment are automatically adjusted according to the turbidity of different regions, thereby realizing the precision and efficiency of data acquisition. This regionalized and parameterized data acquisition strategy not only improves the quality of data, but also significantly shortens the data acquisition cycle;
[0054] The 3D modeling module in the oilfield underwater 3D topography measurement system provided by the present invention adopts advanced image processing and computer vision technology to build an oilfield underwater 3D topography simulation model. By introducing a 3D model abnormal area screening module and a 3D model optimization and updating module, the reliability and real-time performance of the oilfield underwater 3D topography simulation model are significantly enhanced. By verifying the similarity between the 3D reconstructed feature vector parameters of the simulation model and the actual feature vector parameters, the system can quickly identify and screen out abnormal areas. By using multi-source data fusion technology, the abnormal areas are targetedly scanned and updated, thereby ensuring the accuracy and timeliness of the oilfield underwater 3D topography simulation model, and solving the problem in the prior art that the abnormal areas cannot be quickly identified, screened and updated. BRIEF DESCRIPTION OF THE DRAWINGS
[0055] Figure 1 It is a structural block diagram of the oil field underwater three-dimensional topographic measurement system of the present invention;
[0056] Figure 2 The present invention is a flow chart of the method for measuring three-dimensional underwater topography of an oil field. DETAILED DESCRIPTION
[0057] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.
[0058] Example 1, see Figure 1 The invention provides a structural block diagram of an oilfield underwater three-dimensional topographic measurement system, which includes:
[0059] The area division module is used to divide the target underwater oil field into several areas and number them; suppose it is divided into n areas, and i represents the area number;
[0060] A data acquisition equipment optimization module, used to identify the turbidity of each area and output equipment adjustment parameters of the data acquisition equipment based on the relationship between the turbidity and the preset value;
[0061] The underwater oilfield data acquisition module, after adjusting the parameters of the equipment, conducts several rounds of terrain scanning on the underwater oilfield to obtain several sets of three-dimensional information for each area; underwater cameras, sonar, and lidar equipment are used to collect data on the underwater environment; the collected three-dimensional information includes water depth, terrain, and underwater structure;
[0062] The data preprocessing module is used to clean, denoise and correct several groups of 3D information of each underwater area of the oil field, and use image processing and computer vision technologies such as optical flow and stereo matching to extract the 3D features of the underwater scene of each area;
[0063] The 3D modeling module constructs a 3D underwater terrain simulation model of the oil field based on the 3D features of the underwater scene in each area through rotation, scaling, translation and registration operations;
[0064] The 3D model abnormal area screening module is used to verify the similarity between the 3D reconstructed feature vector parameters and the actual feature vector parameters of the underwater 3D terrain simulation model of the oil field, and screen out abnormal areas;
[0065] The 3D model optimization and updating module scans the abnormal area to obtain multi-source data, and replaces the abnormal area of the oilfield underwater 3D terrain simulation model based on the fusion of multi-source data to obtain an updated oilfield underwater 3D terrain simulation model.
[0066] It is necessary to further explain in the embodiment of the present invention that when the turbidity of the i-th area exceeds the preset value, the device adjustment parameter of the acquisition device in the i-th area is output; the device adjustment parameter is obtained in the following manner:
[0067] Step S11, turbidity monitoring and determination: using a turbidity meter or a sensor integrated in a laser radar system to monitor the turbidity of the water body in the i-th area in real time; threshold comparison: comparing the real-time turbidity data with a preset turbidity threshold, if the turbidity exceeds the preset value, proceed to the next step;
[0068] Step S12, evaluating the impact of turbidity exceeding a preset value on three-dimensional accuracy, such as signal attenuation and reduction in signal-to-noise ratio; and determining target parameters that need to be adjusted, such as laser power and scanning wavelength, based on the evaluation results;
[0069] Step S13, laser power adjustment: according to the degree of turbidity exceeding, calculate or select the appropriate increase in laser output power; the constraint condition is that the increase in laser power does not exceed the safety limit and the hardware tolerance; according to the penetration ability and scattering characteristics of lasers of different wavelengths in turbid water bodies, select the appropriate wavelength combination and configure the multi-wavelength laser scanning system to obtain clearer images and more accurate distance measurement information;
[0070] Step S14, output device adjustment parameters: integrate the calculated laser output power increase and multi-wavelength laser scanning configuration; transmit the integrated adjustment parameters to the collection device in the i-th area through the control system; after receiving the adjustment parameters, the collection device adjusts its working state to adapt to the current turbidity level;
[0071] Step S15, monitoring and feedback: After the adjustment parameters are applied, the performance changes of the lidar system are continuously monitored to evaluate the impact of the adjustment parameters on the detection effect in turbid water bodies; if the adjustment effect is not ideal, the parameters are readjusted according to the monitoring results and evaluation feedback until a satisfactory detection effect is achieved.
[0072] It needs to be further explained in the embodiment of the present invention that the method for obtaining the increase in laser output power is:
[0073] The laser output power under standard conditions is obtained, which is recorded as Pba; the laser output power that needs to be increased is Pinc; the total laser output power after adjustment is Pto;
[0074] By formula The signal attenuation factor A caused by turbidity is calculated. The signal attenuation factor A is dimensionless and greater than 1.
[0075] By formula The calculated laser output power that needs to be increased is Pinc;
[0076] By formula The adjusted total laser output power is calculated.
[0077] It needs to be further explained in the embodiment of the present invention that the abnormal area is obtained in the following manner:
[0078] Based on manual measurement, actual measurement data of each area is obtained, and actual feature vector parameters are obtained from the actual measurement data; three-dimensional reconstruction feature vector parameters corresponding to the actual measurement data are obtained from the underwater three-dimensional terrain simulation model of the oil field; j represents the number of the feature vector parameter, and the number of the feature vector parameters is m;
[0079] Based on the actual feature vector parameters and the 3D reconstructed feature vector parameters, the similarity parameters of the feature vector parameters of each region are calculated. The feature similarity parameters of the i-th region are calculated as follows: , where sj represents the jth actual eigenvector parameter, and gj represents the jth 3D reconstructed eigenvector parameter;
[0080] By formula The accuracy index Zp of the underwater three-dimensional terrain simulation model of the oil field is calculated, where c i represents the feature similarity parameter of the i-th region, kc i represents the regional weight coefficient of the ith region. The calculation of the regional weight coefficient satisfies , η1 represents the accuracy influencing factor of manual measurement, η2 represents the preset 3D reconstruction accuracy influencing factor, and 0.95<η1<1.0, 0.85<η2<1.0, which is set based on actual conditions;
[0081] The accuracy index Zp of the oilfield underwater 3D terrain simulation model is compared with the preset value. When the accuracy index Zp is lower than the preset value, it indicates that the quality of the oilfield underwater 3D terrain simulation model is lower than the preset value C. 预, re-collect data and build an underwater three-dimensional terrain simulation model of the oil field; when the accuracy index Zp is not lower than the preset value C 预 , indicating that the quality of the underwater three-dimensional terrain simulation model of the oilfield meets the preset requirements; the characteristic similarity parameters of each area are compared with the preset similarity one by one, and the abnormal areas are screened out; the areas with characteristic similarity parameters lower than the preset similarity are recorded as abnormal areas.
[0082] It is necessary to further explain in the embodiment of the present invention that the number of characteristic vector parameters is set based on the depth and geological instability of the region, and the number of characteristic vector parameters is obtained as follows:
[0083] Assume that the number of characteristic vector parameters required per unit volume in a stable oil field underwater is △C; through the formula The number of characteristic vector parameters is calculated; wherein m represents the number of characteristic vector parameters, V represents the regional volume, QF" represents the geological instability parameter after linear normalization, and SD represents the maximum water depth of the region; the geological instability parameter represents the deformation speed of the underwater topography of the oil field in the region.
[0084] It needs to be further explained in the embodiment of the present invention that the three-dimensional model optimization and updating module includes the following steps:
[0085] Step S21, collecting multi-source data: using high-precision scanning sonar and multi-beam sonar equipment to comprehensively scan the underwater abnormal area of the oil field, the acquired data includes the depth, slope, and morphological three-dimensional parameters of the seabed terrain, forming a first set of three-dimensional terrain data; using an underwater laser radar system to measure the seabed point cloud data by laser pulse reflection; the laser radar data provides a high-density three-dimensional point cloud to form a second set of three-dimensional terrain data; environmental sensor data: using temperature sensors, salinity meters, and flow rate meters deployed underwater to collect water temperature, salinity, and flow rate environmental parameters in real time;
[0086] Step S22, data preprocessing: using timestamps and GPS data to synchronize data from different sensors and time points; using calibration algorithms to convert data from different sources into a unified geographic coordinate system and projection system to achieve spatial calibration; analyzing the error range and stability of each data source under different conditions;
[0087] Step S23, according to the water depth, water flow velocity, sensor accuracy parameters, combined with historical data and theoretical models, fuzzy logic and neural network methods are used for comprehensive evaluation to calculate the positioning accuracy index of each data source; based on the positioning accuracy index, the weighted average method, entropy weight method or hierarchical analysis method are used to calculate the weight coefficient of each data source in different areas and depths;
[0088] Step S24, data fusion and model update: using the calculated weight coefficient, adopting weighted average or Bayesian fusion to fuse the sonar data, lidar data and other sensor data; using the fused three-dimensional reconstruction feature vector parameters to replace the abnormal area data of the oilfield underwater three-dimensional terrain simulation model, to obtain an updated oilfield underwater three-dimensional terrain simulation model;
[0089] Step S25, verifying the accuracy and reliability of the updated oilfield underwater three-dimensional terrain simulation model by comparing it with actual exploration data or other independent data sources.
[0090] It needs to be further explained in the embodiment of the present invention that the positioning accuracy index of the sonar detection data is recorded as SZa, and the positioning accuracy index of the laser radar data is recorded as SZb;
[0091] The three-dimensional reconstruction feature vector parameters of sonar detection data are recorded as SZa_j, and the three-dimensional reconstruction feature vector parameters of lidar data are recorded as SZb_j;
[0092] By formula The calculated 3D reconstruction feature vector parameter is recorded as the weight coefficient of SZa_j; The calculated three-dimensional reconstruction feature vector parameters of the lidar data are recorded as the weight coefficient of SZb_j;
[0093] Based on the weight coefficient, through the formula The fused 3D reconstruction feature vector parameter R_s is calculated.
[0094] It needs to be further explained in the embodiment of the present invention that the positioning accuracy index is obtained in the following manner:
[0095] A convolutional neural network model is constructed, and the water depth, water flow velocity, and sensor accuracy parameters are used as input features. The known positioning accuracy in historical data is used as the label, i.e., the target output. By training the convolutional neural network model, the nonlinear mapping relationship from the input parameters to the positioning accuracy index is learned. After the training is completed, the positioning accuracy index is obtained from the trained neural network model.
[0096] In a further design, the oilfield underwater three-dimensional topographic measurement system further includes:
[0097] The 3D modeling optimization module uses advanced modeling technologies such as NURBS (non-uniform rational B-spline) surface processing and Loft lofting operations to improve the realism and accuracy of the model; it uses texture mapping, material editing and other functions to add color, texture, light and shadow effects to the model to enhance the visual effect.
[0098] Example 2, see Figure 2The present invention provides a method for measuring the three-dimensional underwater topography of an oil field, comprising the following steps:
[0099] Step 1: Divide the target underwater oil field into several areas and number them; divide them into n areas, and use i to represent the area number;
[0100] Step 2: Identify the turbidity of each area, and output the device adjustment parameters of the data acquisition device based on the relationship between the turbidity and the preset value;
[0101] Step 3: After adjusting the parameters of the equipment, several rounds of terrain scanning are carried out on the underwater oil field to obtain several sets of three-dimensional information of each area; underwater cameras, sonar, and lidar equipment are used to collect data on the underwater environment; the collected three-dimensional information includes water depth, terrain, and underwater structure;
[0102] Step 4: Clean, denoise and correct several groups of 3D information of each underwater area of the oil field, and use image processing and computer vision techniques, such as optical flow method and stereo matching, to extract the 3D features of the underwater scene of each area;
[0103] Step 5: construct an underwater 3D terrain simulation model of the oil field based on the 3D features of the underwater scene in each area through rotation, scaling, translation and registration operations;
[0104] Step 6: verify the similarity between the three-dimensional reconstructed feature vector parameters and the actual feature vector parameters of the underwater three-dimensional terrain simulation model of the oil field, and screen out abnormal areas;
[0105] Step 7: Scan the abnormal area to obtain multi-source data, and replace the abnormal area of the oilfield underwater three-dimensional terrain simulation model based on the fusion of the multi-source data to obtain an updated oilfield underwater three-dimensional terrain simulation model.
[0106] The above are only some preferred embodiments of the present invention. Any person skilled in the art may modify the above technical solutions or modify them into equivalent technical solutions. Therefore, the corresponding simple modifications or equivalent transformations made according to the technical solutions of the present invention shall fall within the scope of protection claimed by the present invention.
Claims
1. An underwater three-dimensional topographic measurement system for oil fields, characterized in that: include: A region division module is used to divide the target underwater oil field into several regions and number them; Suppose it is divided into n regions, and i represents the region number; The data acquisition equipment optimization module identifies the turbidity of each area and outputs the equipment adjustment parameters of the data acquisition equipment based on the relationship between the turbidity and the preset value; The underwater oilfield data acquisition module, after adjusting the parameters of the equipment, conducts several rounds of terrain scanning on the underwater oilfield to obtain several sets of three-dimensional information for each area; underwater cameras, sonar, and lidar equipment are used to collect data on the underwater environment; the collected three-dimensional information includes water depth, terrain, and underwater structure; The data preprocessing module is used to clean, denoise and correct several groups of 3D information of each underwater area of the oil field, and use image processing and computer vision technology to extract the 3D features of the underwater scene of each area; The 3D modeling module constructs a 3D underwater terrain simulation model of the oil field based on the 3D features of the underwater scene in each area through rotation, scaling, translation and registration operations; The 3D model abnormal area screening module is used to verify the similarity between the 3D reconstructed feature vector parameters and the actual feature vector parameters of the underwater 3D terrain simulation model of the oil field, and screen out abnormal areas; The 3D model optimization and updating module scans the abnormal area to obtain multi-source data, and replaces the abnormal area of the oilfield underwater 3D terrain simulation model based on the fusion of multi-source data to obtain an updated oilfield underwater 3D terrain simulation model; The three-dimensional model optimization and updating module comprises the following steps: Step S21, collecting multi-source data: using high-precision scanning sonar and multi-beam sonar equipment to comprehensively scan the underwater abnormal area of the oil field, the acquired data includes the depth, slope, and morphological three-dimensional parameters of the seabed terrain, forming a first set of three-dimensional terrain data; using an underwater laser radar system to measure the seabed point cloud data by laser pulse reflection; the laser radar data provides a high-density three-dimensional point cloud to form a second set of three-dimensional terrain data; environmental sensor data: using temperature sensors, salinity meters, and flow rate meters deployed underwater to collect water temperature, salinity, and flow rate environmental parameters in real time; Step S22, data preprocessing: using timestamps and GPS data to synchronize data from different sensors and time points; using calibration algorithms to convert data from different sources into a unified geographic coordinate system and projection system to achieve spatial calibration; analyzing the error range and stability of each data source under different conditions; Step S23, according to the water depth, water flow velocity, sensor accuracy parameters, combined with historical data and theoretical models, fuzzy logic and neural network methods are used for comprehensive evaluation to calculate the positioning accuracy index of each data source; based on the positioning accuracy index, the weighted average method, entropy weight method or hierarchical analysis method are used to calculate the weight coefficient of each data source in different areas and depths; Step S24, data fusion and model update: using the calculated weight coefficient, adopting weighted average or Bayesian fusion to fuse the sonar data, lidar data and other sensor data; using the fused three-dimensional reconstruction feature vector parameters to replace the abnormal area data of the oilfield underwater three-dimensional terrain simulation model, to obtain an updated oilfield underwater three-dimensional terrain simulation model; Step S25, verifying the accuracy and reliability of the updated oilfield underwater three-dimensional terrain simulation model by comparing it with actual exploration data or other independent data sources; The positioning accuracy index of sonar detection data is recorded as SZa, and the positioning accuracy index of lidar data is recorded as SZb; The three-dimensional reconstruction feature vector parameters of sonar detection data are recorded as SZa_j, and the three-dimensional reconstruction feature vector parameters of lidar data are recorded as SZb_j; By formula The calculated 3D reconstruction feature vector parameter is recorded as the weight coefficient of SZa_j; The calculated three-dimensional reconstruction feature vector parameters of the lidar data are recorded as the weight coefficient of SZb_j; Based on the weight coefficient, through the formula Calculate and obtain the fused three-dimensional reconstruction feature vector parameter R_s; The positioning accuracy index is obtained as follows: A convolutional neural network model is constructed, and the water depth, water flow velocity, and sensor accuracy parameters are used as input features. The known positioning accuracy in historical data is used as the label, i.e., the target output. By training the convolutional neural network model, the nonlinear mapping relationship from the input parameters to the positioning accuracy index is learned. After the training is completed, the positioning accuracy index is obtained from the trained neural network model.
2. The oilfield underwater three-dimensional topographic measurement system according to claim 1, characterized in that: The laser output power increase is obtained as follows: Obtain the laser output power under standard conditions, recorded as Pba; the laser output power that needs to be increased is Pinc; the adjusted total laser output power is Pto; through the formula The signal attenuation factor A caused by turbidity is calculated. The signal attenuation factor A is dimensionless and greater than 1. By formula The calculated laser output power that needs to be increased is Pinc; By formula The adjusted total laser output power is calculated.
3. The oilfield underwater three-dimensional topographic measurement system according to claim 1, characterized in that: The abnormal area is obtained in the following manner: Based on manual measurement, actual measurement data of each area is obtained, and actual feature vector parameters are obtained from the actual measurement data; three-dimensional reconstruction feature vector parameters corresponding to the actual measurement data are obtained from the underwater three-dimensional terrain simulation model of the oil field; j represents the number of the feature vector parameter, and the number of the feature vector parameters is m; Based on the actual feature vector parameters and the 3D reconstructed feature vector parameters, the similarity parameters of the feature vector parameters of each region are calculated. The feature similarity parameters of the i-th region are calculated as follows: , where sj represents the jth actual eigenvector parameter, and gj represents the jth 3D reconstructed eigenvector parameter; By formula The accuracy index Zp of the underwater three-dimensional terrain simulation model of the oil field is calculated, where c i represents the feature similarity parameter of the i-th region, kc i represents the regional weight coefficient of the ith region. The calculation of the regional weight coefficient satisfies , η1 represents the accuracy influencing factor of manual measurement, η2 represents the preset 3D reconstruction accuracy influencing factor, and 0.95<η1<1.0, 0.85<η2<1.0, which is set based on actual conditions; The accuracy index Zp of the oilfield underwater 3D terrain simulation model is compared with the preset value. When the accuracy index Zp is lower than the preset value, it indicates that the quality of the oilfield underwater 3D terrain simulation model is lower than the preset value C. 预 , re-collect data and build an underwater three-dimensional terrain simulation model of the oil field; when the accuracy index Zp is not lower than the preset value C 预 , indicating that the quality of the underwater three-dimensional terrain simulation model of the oilfield meets the preset requirements; the characteristic similarity parameters of each area are compared with the preset similarity one by one, and the abnormal areas are screened out; the areas with characteristic similarity parameters lower than the preset similarity are recorded as abnormal areas.
4. The oilfield underwater three-dimensional topographic measurement system according to claim 3, characterized in that: The number of eigenvector parameters is set based on the depth and geological instability of the region. The number of eigenvector parameters is obtained as follows: Assume that the number of characteristic vector parameters required per unit volume in a stable oil field underwater is △C; through the formula Calculate the number of eigenvector parameters; Where m represents the number of eigenvector parameters, V represents the regional volume, QF" represents the geological instability parameter after linear normalization, and SD represents the maximum water depth of the region; the geological instability parameter represents the deformation speed of the underwater topography of the oil field in the region.
5. A measurement method using the oilfield underwater three-dimensional topographic measurement system according to any one of claims 1 to 4, characterized in that: The following steps are involved: Step 1: Divide the target underwater oil field into several areas and number them; divide them into n areas, and use i to represent the area number; Step 2: Identify the turbidity of each area, and output the device adjustment parameters of the data acquisition device based on the relationship between the turbidity and the preset value; Step 3: After adjusting the parameters of the equipment, several rounds of terrain scanning are carried out on the underwater oil field to obtain several sets of three-dimensional information of each area; underwater cameras, sonar, and lidar equipment are used to collect data on the underwater environment; the collected three-dimensional information includes water depth, terrain, and underwater structure; Step 4: Clean, denoise and correct several groups of 3D information of each underwater area of the oil field, and use image processing and computer vision techniques, such as optical flow method and stereo matching, to extract the 3D features of the underwater scene of each area; Step 5: construct an underwater 3D terrain simulation model of the oil field based on the 3D features of the underwater scene in each area through rotation, scaling, translation and registration operations; Step 6: verify the similarity between the three-dimensional reconstructed feature vector parameters and the actual feature vector parameters of the underwater three-dimensional terrain simulation model of the oil field, and screen out abnormal areas; Step 7: Scan the abnormal area to obtain multi-source data, and replace the abnormal area of the oilfield underwater three-dimensional terrain simulation model based on the fusion of the multi-source data to obtain an updated oilfield underwater three-dimensional terrain simulation model.
