A lightweight oil and gas exploration system and dynamic exploration method

Abstract

This invention discloses a lightweight oil and gas exploration system and a dynamic exploration method, belonging to the field of oil and gas exploration technology. This invention utilizes modular lightweight exploration units in conjunction with relay UAVs to completely break the limitations of complex terrain on equipment deployment. Compared with traditional exploration methods, deployment efficiency is greatly improved, and the operability of exploration work in complex environments is enhanced. The application of a dynamic density adjustment mechanism can intelligently reduce the amount of invalid data collection based on real-time data feedback, significantly reducing exploration costs while ensuring exploration effectiveness.

Description

Technical Field

[0001] This invention relates to the field of oil and gas exploration technology, and in particular to a lightweight oil and gas exploration system and dynamic exploration method. Background Technology

[0002] Oil and natural gas are important strategic resources for my country. As the first key link in the exploitation of oil and gas resources, the development and improvement of oil and gas exploration technology is a key focus of research for researchers in the field.

[0003] In oil and gas exploration, seismic wave exploration is one of the mainstream methods, but it has many undeniable drawbacks. Its reliance on artificial seismic sources and fixed geophone arrays results in extremely bulky equipment. In complex terrain environments such as mountains and jungles, the transportation and deployment of this equipment are extremely difficult, significantly limiting the scope and efficiency of exploration work. Furthermore, this traditional method offers relatively limited data dimensions, relying solely on seismic wave parameters for analysis, making it highly susceptible to noise interference and severely impacting the accuracy of reservoir prediction. Moreover, the complexity of the surface environment exponentially increases the difficulty of deploying geophone arrays, and traditional data processing models heavily depend on backend servers, resulting in severely insufficient data processing timeliness, failing to meet the demands of today's rapid exploration.

[0004] Based on the above, a lightweight oil and gas exploration system and dynamic exploration method are proposed. Summary of the Invention

[0005] The purpose of this invention is to provide a lightweight oil and gas exploration system and dynamic exploration method to solve the problems in the background art.

[0006] To achieve the above objectives, the present invention provides a lightweight oil and gas exploration system, comprising multiple modular lightweight exploration units, an intelligent relay UAV unit, a dynamic networking system, and a ground control center. The modular lightweight exploration units are connected to the intelligent relay UAV unit through the dynamic networking system. The intelligent relay UAV unit receives signals transmitted by the modular lightweight exploration units and provides feedback. The ground control center is wirelessly connected to both the intelligent relay UAV unit and the modular lightweight units.

[0007] Preferably, the modular lightweight exploration unit is composed of multiple exploration units. Each exploration unit includes a unit body and a positioning and navigation component disposed inside the unit body, and a drive control component disposed at the lower end of the unit body. The positioning and navigation component is configured as a GNSS module, and the drive control component includes a moving mechanism and an adaptive anchoring mechanism disposed in the middle of the moving mechanism.

[0008] Preferably, the unit body is made of carbon fiber composite material, a triaxial seismic detector is provided at the center of the unit body, an electromagnetic field sensor is provided on one side of the unit body, a miniature gravimeter is provided at the bottom of the unit body, a temperature sensor is provided at the top of the unit body, and an edge computing module is provided inside the unit body.

[0009] Preferably, the intelligent relay drone unit is configured as a rotary-wing drone equipped with a magnetic anomaly detector, and the rotary-wing drone is equipped with a data analysis module, which employs AI algorithms and deep learning algorithms.

[0010] Preferably, the dynamic networking system is a LoRa+Mesh self-organizing network, and both the main unit and the intelligent relay UAV unit are equipped with a LoRa+Mesh self-organizing network communication module.

[0011] Based on the aforementioned lightweight oil and gas exploration system, this invention proposes a dynamic exploration method, comprising the following steps:

[0012] S1. Deploy exploration units and establish communication connections. Activate the triaxial seismic detectors, electromagnetic field sensors, miniature gravimeters, and temperature and pressure sensors on each exploration unit. Continuously collect relevant raw data according to the set sampling frequency, and process the relevant data in real time using the edge computing module.

[0013] S2. Simultaneously with the activation of each exploration unit, the rotary-wing intelligent relay drone begins its first round of patrols along a preset path, receiving the processed raw data from each exploration unit in real time, initially constructing a basic data framework for the exploration area, and using the built-in data analysis module to analyze the basic data, identify areas where oil and gas anomalies may exist, and calculate the optimal path, new deployment spacing, and migration time plan for the surrounding exploration units to migrate to the anomaly area based on the size, shape, location of the anomaly area and the situation of the surrounding deployed exploration units, generating an equipment redeployment plan and issuing migration instructions to the relevant exploration units.

[0014] S3. After receiving the migration command, the relevant exploration unit activates the bottom adaptive anchoring mechanism to release the fixation, moves to the new position according to the planned path and re-fixes itself. After the migration is completed, the exploration unit starts the high-frequency acquisition mode to collect data in the abnormal area. At the same time, the edge node module is used to perform local inversion calculation on the collected data and extract abnormal feature data. The abnormal feature data includes seismic wave characteristics and abnormal electromagnetic signal characteristics within a specific frequency range. Then, these abnormal feature data are uploaded to the intelligent relay drone.

[0015] S4. The intelligent relay UAV unit will transmit the abnormal feature data received from each exploration unit and the magnetic anomaly data collected by its own magnetic anomaly detector to the ground control center.

[0016] S5. Construct a DC-GNN model at the ground control center, deeply fusing multiphysics data with historical exploration databases. Use the fused data to train and optimize the DC-GNN model, adjusting model parameters such as convolutional kernel size, number of neural network layers, and learning rate to improve the model's ability to identify reservoir characteristics. Cross-validation is used during training to evaluate model performance.

[0017] Among them, the historical exploration database is organized and labeled in advance. The historical exploration database includes relevant data characteristics of known oil and gas reservoirs under different geological conditions.

[0018] S6. The trained DC-GNN model is used to predict and analyze the data of the current exploration area, and outputs an oil and gas probability cloud map. Exploration personnel analyze the oil and gas probability cloud map and select exploration targets.

[0019] Preferably, the step of deploying exploration units and establishing communication connections in S1 is as follows: based on the geological data of the exploration area, plan the initial distribution scheme of modular lightweight exploration units and the preset flight path of intelligent relay drones, deploy and start the exploration units according to the initial distribution scheme, and establish communication connections between each exploration unit through LoRa+Mesh self-organizing network.

[0020] Preferably, in S1, the relevant raw data includes seismic wave velocity, electromagnetic impedance, gravity gradient, and geothermal field data;

[0021] The edge computing module uses wavelet transform combined with Kalman filtering to eliminate noise in the acquired raw data. Specifically, the acquired raw data is first subjected to wavelet transform to decompose different frequency components, identify and filter out noise frequency band signals, and then Kalman filtering is used to dynamically estimate and optimize the remaining signal.

[0022] Preferably, in step S5, the multiphysics data includes seismic wave velocity, electromagnetic impedance, gravity gradient, geothermal field, and magnetic anomaly data.

[0023] Therefore, the lightweight oil and gas exploration system and dynamic exploration method of the present invention have the following beneficial effects:

[0024] (1) In this invention, the modular lightweight exploration unit is used in conjunction with the relay UAV to completely break the limitation of equipment deployment on complex terrain. Compared with the traditional exploration method, the deployment efficiency is greatly improved, and the operability of exploration work in complex environments is improved. At the same time, the three-axis seismic detector, electromagnetic field sensor, micro gravimeter and temperature and pressure sensor are used to detect and obtain multi-physics field data, which effectively eliminates the interference between the components, improves the signal-to-noise ratio, and significantly improves the quality of data acquisition.

[0025] (2) The present invention adopts a multi-physics spatiotemporal joint inversion algorithm to comprehensively analyze multiple physical field data in time and space dimensions, which greatly improves the accuracy of reservoir identification and provides more reliable technical support for oil and gas exploration. The application of its dynamic density adjustment mechanism can intelligently reduce the amount of invalid data collection based on real-time data feedback, which significantly reduces exploration costs while ensuring exploration results.

[0026] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0027] Figure 1 This is a cross-sectional view of the exploration unit according to an embodiment of the present invention;

[0028] Figure 2 This is a flowchart of the exploration method according to an embodiment of the present invention;

[0029] Figure label:

[0030] 1. Unit body; 2. Electromagnetic field sensor; 3. Triaxial seismic detector; 4. Miniature gravimeter; 5. Adaptive anchoring mechanism; 6. Moving mechanism; 7. GNSS module; 8. Edge computing module; 9. Temperature and pressure sensor; 10. Control system. Detailed Implementation

[0031] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0032] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0033] Example

[0034] This invention provides a lightweight oil and gas exploration system, comprising multiple modular lightweight exploration units, an intelligent relay UAV unit, a dynamic networking system, and a ground control center. The modular lightweight exploration units and the intelligent relay UAV unit are connected via a dynamic networking system, which is a LoRa+Mesh self-organizing network. Both the modular lightweight exploration units and the intelligent relay UAV unit are equipped with LoRa+Mesh self-organizing network communication modules. The ground control system is wirelessly connected to the modular lightweight exploration units and the intelligent relay UAV unit, and a remote monitoring system is established using a satellite communication link.

[0035] The intelligent relay UAV unit receives signals from the modular lightweight exploration unit and provides feedback. The ground control center is wirelessly connected to both the intelligent relay UAV unit and the modular lightweight unit.

[0036] like Figure 1 As shown, the modular lightweight exploration unit consists of multiple exploration units. Each exploration unit includes a control system 10, a unit body 1, a positioning and navigation component located inside the unit body 1, and a drive control component located at the lower end of the unit body 1. The control system 10 can be located inside the unit body 1. The positioning and navigation component is a GNSS module 7, which provides accurate geographic location information to accurately move according to the direction and position indicated by the intelligent relay UAV signal. The drive control component includes a moving mechanism 6 and an adaptive anchoring mechanism 5 located in the middle of the moving mechanism 6. In this embodiment, the moving mechanism adopts a conventional wheeled, tracked, or legged moving mechanism. When it receives a signal containing movement commands sent by the intelligent relay UAV, the control system 10 of the exploration unit will parse the direction, speed, and other information of movement according to the commands, and then drive the corresponding moving mechanism 6 to move. Different moving mechanisms can be selected according to the actual exploration situation. Figure 1 The system employs a wheeled mobility mechanism, with the motor not shown, and its connection method follows the conventional setup of existing technologies. The wheeled mobility mechanism uses a motor to drive the wheels to rotate, enabling forward, backward, and turning movements; the tracked mobility mechanism moves by driving the tracks, a method more suitable for navigating complex terrain; the legged mobility mechanism achieves walking by controlling the movement of the leg joints, possessing good obstacle-crossing capabilities.

[0037] The adaptive anchoring mechanism 5 is designed to adapt to different surface conditions. The adaptive anchoring mechanism is configured as a retractable spiral anchor or vacuum suction disc driven by a drive motor. Figure 1 The image shows a vacuum adsorption plate; both of these setups utilize conventional structures. In rocky terrain, the retractable spiral ground stakes can quickly extend and screw into rock crevices for secure fixation; while on soft ground, the vacuum adsorption plate comes into play, using its powerful adsorption force to quickly fix the equipment within 5 seconds, ensuring stable operation in various complex terrains.

[0038] The main unit 1 is made of carbon fiber composite material, which has an excellent strength-to-weight ratio, allowing the entire unit to weigh less than 5 kg while ensuring good structural stability. A triaxial seismic detector 3 is located at the center of the main unit 1 to detect seismic wave velocity. A temperature and pressure sensor 9 is located at the top of the main unit to record ground temperature and pressure, and to detect geothermal field data. An electromagnetic field sensor 2 is located on one side of the main unit 1 to measure the electromagnetic impedance of the ground. A miniature gravimeter 4 is located at the bottom of the main unit 1 to monitor changes in gravity gradient, enabling simultaneous acquisition of multiple key parameters. An edge computing module 8 is located inside the main unit 1. The connection routes between the above components are not shown in the diagram. Figure 1 As shown, all connections use conventional methods.

[0039] In addition, this embodiment installs a solar-kinetic dual-mode power supply system on the modular lightweight exploration unit. A small solar panel is fixedly installed on one side of the top of the unit body to ensure efficient reception of light energy under different angles of illumination. A battery is installed inside the unit body and connected to other components. A kinetic energy harvesting device is installed on the mobile mechanism to test the kinetic energy harvesting efficiency at different movement speeds, ensuring that the endurance meets the design requirement of more than 30 days. This device converts and connects to the battery for power generation. Both of these setups adopt conventional structural configurations in the prior art.

[0040] The intelligent relay drone unit is configured as a rotary-wing drone equipped with a magnetic anomaly detector. The magnetic anomaly detector is deployed on the intelligent relay drone to detect magnetic anomalies in the strata. The rotary-wing drone is equipped with a data analysis module that uses reinforcement learning algorithms to analyze the received data and determine the redistribution scheme.

[0041] Based on the aforementioned lightweight oil and gas exploration system, dynamic oil and gas exploration is conducted in the Tarim Basin, such as... Figure 2 As shown, the specific steps are as follows:

[0042] (1) Geological data collection and analysis before exploration:

[0043] The basin contains multiple large folds and fault zones. Studies of stratigraphic lithology have revealed the development of several oil and gas-bearing strata, including Paleozoic marine carbonate rocks and Mesozoic terrestrial clastic rocks. Integrating previous exploration findings, the distribution patterns and geological characteristics of some discovered oil and gas fields have been clarified. Using Geographic Information System (GIS) technology, combined with topographic data of the vast desert and Gobi regions of the Tarim Basin, a preliminary assessment indicates that the piedmont tectonic belts on the basin's edge and the ancient uplift areas within the basin may be favorable areas for oil and gas reservoir development.

[0044] (2) Customization, assembly, and debugging of exploration systems:

[0045] Given the high temperature, aridity, and windy, sandy environment of the Tarim Basin, the modular, lightweight exploration unit was specifically designed. To enhance mobility, the mobile mechanism was upgraded with wide, wear-resistant sand-specific tires, coupled with an optimized suspension system, improving its ability to traverse soft terrain such as sand. Dust protection was strengthened, with high-performance sealant used to seal gaps in all components, and dust covers added to internal electronic components. Custom-designed high-efficiency solar panels were used, achieving a conversion efficiency of over 30% under the basin's intense sunlight conditions. Simultaneously, the kinetic energy harvesting device was optimized to utilize the vibrations and bumps during the exploration unit's movement, more efficiently converting mechanical energy into electrical energy, ensuring stable operation of the equipment in the basin environment.

[0046] At a well-protected base located on the edge of the basin, modular lightweight exploration units were assembled according to high-precision standards. Triaxial seismic detectors, electromagnetic field sensors, microgravimeters, and temperature and pressure sensors were calibrated multiple times using professional calibration equipment. The complex geological signals of the Tarim Basin were simulated, including weak seismic wave signals from deep strata, anomalous electromagnetic signals caused by geological structures, and common wind and sand noise within the basin, to test the built-in edge computing module. Data compression rates were ensured to be consistently above 85%, and the effective signal retention rate after noise filtering reached 90%. Twenty fixation tests were conducted on the bottom adaptive anchoring mechanism, targeting typical terrains such as rock and sand within the basin. In sandy environments, the vacuum adsorption plate quickly adsorbed and stably fixed the mechanism; in rocky areas, the retractable spiral anchors quickly screwed into rock crevices, ensuring stable fixation within 5 seconds in both cases.

[0047] Given the vast area (approximately 560,000 square kilometers) and complex terrain of the Tarim Basin, LoRa+Mesh ad hoc network communication parameters were planned. Communication tests were conducted in different terrain areas within the basin, such as the desert hinterland, piedmont hills, and Gobi desert. The number of exploration units was gradually increased to 200, and communication signal strength was monitored. In the desert hinterland, due to sand dune obstruction, antenna height and angle were adjusted to enhance signal transmission capabilities, ensuring signal strength remained above -80dBm and a transmission rate of 1Mbps. Dynamic topology adjustments could be completed within 5 seconds to adapt to the constantly changing positional relationships of exploration units within the basin, optimizing the communication scheme to ensure stable data transmission.

[0048] To address the strong winds and high temperatures characteristic of the Tarim Basin, the intelligent relay drone underwent protective measures and performance optimization. A high-temperature resistant coating was added to the airframe, and rotors and motors with enhanced wind resistance were selected. A magnetic anomaly detector was precisely installed, and field tests were conducted in known magnetic anomaly areas within the basin, such as around igneous intrusions in the Kuqa region, to ensure measurement accuracy of ±5nT. The automatic charging function of the intelligent relay drone was repeatedly tested in the basin environment. By optimizing the positioning system of the charging base station and the navigation algorithm of the intelligent relay drone, a charging success rate of over 98% was guaranteed. Based on the basin's geographical features and exploration priorities, a pre-planned cruise route was developed, and five simulated flights were conducted, avoiding no-fly zones such as military control areas and nature reserves within the basin, ensuring stable data reception and controlling flight trajectory deviation within ±5m.

[0049] (3) Conduct oil and gas exploration:

[0050] 1) Based on collected geological data and GIS analysis results, key exploration areas were delineated in the Tarim Basin. In geologically complex areas such as the Kuqa Depression, modular lightweight exploration units were deployed using a 150m grid layout to capture geological information more precisely; in relatively simple areas such as the Tazhong Uplift, the spacing was set at 250m. A high-precision GPS positioning system, combined with existing ground control points, was used to accurately plan the deployment location of each exploration unit. In desert areas, pre-set landmarks and satellite imagery were used to ensure that the location error was within ±3m, laying the foundation for the accuracy of subsequent data acquisition.

[0051] A combination of helicopter slingshot and all-terrain vehicle (ATV) transport was used to deliver exploration units to their designated locations. In the heart of the desert, helicopter slingshot deployment allowed for rapid deployment of exploration units to designated locations; in relatively flat areas at the edge of basins, ATV transport was more cost-effective. After deployment, each exploration unit automatically activated and quickly established communication connections with surrounding units via a LoRa+Mesh ad hoc network.

[0052] Each modular lightweight exploration unit, operating at optimized sampling frequencies (1200Hz for seismic detectors and 120Hz for other sensors), synchronously and continuously acquires data on seismic wave velocity, electromagnetic impedance, gravity gradient, and geothermal field. During acquisition, a remote monitoring system is established using a satellite communication link to monitor the real-time operating status of each sensor, including parameters such as temperature and voltage. Data processing is performed through an edge computing module, which employs wavelet transform combined with Kalman filtering to eliminate noise in real time. Targeting the wind and sand noise spectrum characteristics of the Tarim Basin, the decomposition level and threshold of the wavelet transform are optimized to improve noise elimination effectiveness. The processed data is transmitted to the intelligent relay UAV unit in temporal and spatial order; simultaneously, data is backed up daily to the ground control center with a storage capacity of 1TB to ensure data security and prevent data loss due to equipment failure or natural disasters.

[0053] 2) At the same time as each exploration unit is launched, the intelligent relay drone begins its first round of patrols according to the preset path, receives the initial data of each unit in real time, and initially constructs the geological data framework of the Tarim Basin. At the same time, the remote monitoring system is used to monitor the data transmission quality in real time, and promptly detects and resolves problems such as signal interruption.

[0054] After receiving preliminary data from various exploration units during its initial patrol, the intelligent relay drone used AI algorithms optimized specifically for the geological characteristics of the Tarim Basin to conduct in-depth analysis of the data. Combining this with known geological structures within the basin, such as geological features near fault zones and hydrocarbon indications, the drone identified areas with potential hydrocarbon anomalies by comparing the changing patterns of electromagnetic impedance and gravity gradient data. In a certain area of ​​the Kuqa Depression, data analysis revealed a region with a sharp decrease in electromagnetic impedance and an abnormal change in gravity gradient, which was determined to be a potential hydrocarbon anomaly area with boundary accuracy controlled within ±10m.

[0055] Based on the identified anomalous areas, the intelligent relay drone uses reinforcement learning algorithms to comprehensively consider the topography of the Tarim Basin, such as dune height and orientation, mountain distribution, as well as the location of already deployed units and data collection needs, to generate a detailed equipment redeployment plan. It calculates the optimal path for surrounding units to migrate to the anomalous area, taking into account the impact of sand dune movement and soft sand on equipment movement in the desert region, and plans the path to avoid large dunes and sand pits as much as possible. The new deployment spacing is determined to be 40m, and the migration time is precisely planned, choosing the early morning when winds are relatively weak to ensure an efficient and orderly migration process, preventing equipment from getting stuck in the sand or being affected by strong winds during the migration.

[0056] 3) Upon receiving the migration command, the relevant exploration units quickly activate their bottom adaptive anchoring mechanisms to release their anchors, move to the new location according to the planned path, and re-anchor themselves. During the migration, a high-precision real-time positioning system and a stable communication link are used to ensure that the units accurately reach the designated location, with positional deviations controlled within ±2m. After migration, the exploration units initiate a high-frequency acquisition mode, increasing the sampling frequency of the seismic detectors to 2500Hz and the sampling frequency of other sensors to 250Hz, enabling more refined data acquisition of anomaly areas. Simultaneously, edge nodes utilize local computing resources to perform localized inversion calculations on the acquired data, extracting seismic wave characteristics related to oil and gas reservoirs in the Tarim Basin under specific geological structures, such as seismic wave reflection characteristics of carbonate reservoirs, wave impedance characteristics of clastic reservoirs, and anomalous electromagnetic signal characteristics. These anomalous feature data are then uploaded to the intelligent relay drone, reducing bandwidth consumption while ensuring data validity.

[0057] 4) Intelligent relay: The intelligent relay UAV will transmit the abnormal feature data collected from each exploration unit and the data collected by its own magnetic anomaly detector to the ground control center located at the edge of the basin via satellite communication link.

[0058] 5) At the ground control center, a deep convolutional graph neural network (DC-GNN) model specifically designed for the geological data characteristics of the Tarim Basin was constructed. The multiphysics field data was deeply integrated with the historical exploration database of the Tarim Basin. The historical exploration database was updated to include the latest seismic, well logging and other exploration data, as well as the geological understanding of newly discovered oil and gas reservoirs in the basin, so that the model is more in line with the actual geological conditions of the basin.

[0059] The DC-GNN model was trained and optimized using the fused data. Based on the characteristics of the geological data in the Tarim Basin, such as the non-uniformity of data distribution and the differences in data features under different geological structures, the model parameters were adjusted. The convolution kernel size was adjusted to 3×3 to better extract local features, and the number of neural network layers was increased to 10 to improve the model's ability to learn complex geological features. The learning rate was optimized to 0.001. Cross-validation and leave-one-out methods were used, combined with validation data from known oil and gas reservoirs in the Tarim Basin, to evaluate the model's performance. After multiple rounds of training and optimization, the model's accuracy in identifying reservoir features in the Tarim Basin reached over 92%.

[0060] 6) After training, the model is used to analyze and predict current exploration data in the Tarim Basin, outputting a probability cloud map of hydrocarbon-bearing areas with a resolution of 10m×10m×5m (XYZ direction). The probability cloud map's color scheme and labeling are optimized, using warm colors such as red, orange, and yellow to represent high-probability areas and cool colors such as blue and green to represent low-probability areas, with color intervals corresponding to different probability values. The model predicts that the favorable reservoir depth range for hydrocarbon-bearing areas in this region is 6700-7300 meters, with sandstone as the main lithology, porosity between 15% and 20%, and permeability between 40 and 60 millidarcy. The model predicts that high-abundance hydrocarbons may exist at around 7000 meters.

[0061] Exploration personnel combined their geological knowledge of the Tarim Basin, such as stratigraphic sedimentary environment and tectonic evolution history, with practical experience to conduct in-depth analysis of probability cloud maps. In a certain area of ​​the Tarim Uplift, through analysis of probability cloud maps, high-probability areas with probability values ​​>90% were selected as key exploration targets. At the same time, low-probability areas were reasonably evaluated to exclude areas with false anomalies due to complex geological structures, thereby reducing exploration risks.

[0062] The above exploration results were verified and compared. Based on the reservoir probability cloud map, areas with higher probability values ​​were selected. Here, an area with a probability value >90% in the Tazhong Uplift was selected as the key target area, and a professional drilling team was arranged to carry out drilling verification work.

[0063] During the drilling process, detailed drilling data were recorded. The drilling depth in this area reached 7500 meters. Core samples encountered showed that the 6800-7200 meter depth range consisted of Mesozoic terrestrial clastic rocks, primarily interbedded sandstone and mudstone. The sandstone porosity, as measured in the laboratory, averaged 18%, with a permeability of 50 millidarcy, indicating good reservoir potential. Significant oil and gas shows were detected at depths of 7000-7050 meters. Oil traces were observed on the core surface, and fluorescence detection showed a strong positive result. Gas logging revealed a significant increase in hydrocarbon content at this depth, with methane content reaching a maximum of 85%, and a certain proportion of heavy hydrocarbons such as ethane and propane.

[0064] A comparative analysis of drilling data and DC-GNN model predictions revealed that the model predicted favorable reservoir depths of 6700-7300 meters in this area, which was consistent with the actual drilling findings. The predicted depth error was ±50 meters, within an acceptable range. The model predicted that the reservoir lithology was predominantly sandstone, with porosity between 15% and 20% and permeability between 40 and 60 millidarcy, largely consistent with actual core measurements. Regarding the location of oil and gas shows, the model predicted high-abundance oil and gas at approximately 7000 meters. Actual drilling at this depth indeed revealed significant oil and gas shows, further validating the effectiveness of the model's predictions. This demonstrates that the exploration method possesses high accuracy and reliability in the Tarim Basin.

[0065] Therefore, the present invention provides a lightweight oil and gas exploration system and dynamic exploration method that utilizes modular lightweight exploration units in conjunction with relay drones to completely break the limitations imposed by complex terrain on equipment deployment. Compared with traditional exploration methods, deployment efficiency is significantly improved, enhancing the operability of exploration work in complex environments. The application of a dynamic density adjustment mechanism can intelligently reduce the amount of invalid data collection based on real-time data feedback, significantly reducing exploration costs while ensuring exploration effectiveness.

[0066] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A lightweight oil and gas exploration system, characterized in that: The system comprises multiple modular lightweight exploration units, intelligent relay UAV units, a dynamic networking system, and a ground control center. The modular lightweight exploration units and the intelligent relay UAV units are connected via the dynamic networking system. The intelligent relay UAV units receive signals emitted by the modular lightweight exploration units and provide feedback. The ground control center is wirelessly connected to both the intelligent relay UAV units and the modular lightweight exploration units. Each modular lightweight exploration unit consists of multiple exploration units, including a unit body, a positioning and navigation component inside the unit body, and a drive control component located at the lower end of the unit body. The positioning and navigation component is a GNSS module, and the drive control component includes a movement mechanism and an adaptive anchoring mechanism located in the middle of the movement mechanism. The intelligent relay UAV unit is a rotary-wing UAV equipped with a magnetic anomaly detector and a data analysis module. The dynamic networking system is a LoRa+Mesh self-organizing network, and both the unit body and the intelligent relay UAV units have LoRa+Mesh self-organizing network communication modules inside. A dynamic exploration method based on a lightweight oil and gas exploration system includes the following steps: S1. Deploy exploration units and establish communication connections. Turn on the triaxial seismic detectors, electromagnetic field sensors, miniature gravimeters and temperature and pressure sensors on each exploration unit. Continuously collect relevant raw data according to the set sampling frequency and process the relevant data in real time using the edge computing module. S2. Simultaneously with the activation of each exploration unit, the rotary-wing intelligent relay drone begins its first round of patrols along a preset path, receiving the processed raw data from each exploration unit in real time, initially constructing a basic data framework for the exploration area, and using the built-in data analysis module to analyze the basic data, identify areas where oil and gas anomalies may exist, and calculate the optimal path, new deployment spacing, and migration time plan for the surrounding exploration units to migrate to the anomaly area based on the size, shape, location of the anomaly area and the situation of the surrounding deployed exploration units, generating an equipment redeployment plan and issuing migration instructions to the relevant exploration units. S3. After receiving the migration command, the relevant exploration unit activates the bottom adaptive anchoring mechanism to release the fixation, moves to the new position according to the planned path and re-fixes itself. After the migration is completed, the exploration unit starts the high-frequency acquisition mode to collect data in the abnormal area. At the same time, the edge node module is used to perform local inversion calculation on the collected data to extract abnormal feature data. The abnormal feature data includes seismic wave characteristics and abnormal electromagnetic signal characteristics within a specific frequency range. Then, these abnormal feature data are uploaded to the intelligent relay drone. S4. The intelligent relay UAV unit will transmit the abnormal feature data received from each exploration unit and the magnetic anomaly data collected by its own magnetic anomaly detector to the ground control center. S5. Construct a DC-GNN model at the ground control center, deeply integrate multiphysics data with historical exploration databases, use the integrated data to train and optimize the DC-GNN model, adjust model parameters, and use cross-validation to evaluate model performance; in this process, the historical exploration database is organized and labeled in advance, and includes relevant data features of known oil and gas reservoirs under different geological conditions. S6. The trained DC-GNN model is used to predict and analyze the data of the current exploration area, and outputs an oil and gas probability cloud map. Exploration personnel analyze the oil and gas probability cloud map and select exploration targets.

2. The lightweight oil and gas exploration system according to claim 1, characterized in that: The main body of the unit is made of carbon fiber composite material. A triaxial seismic detector is installed at the center of the main body. An electromagnetic field sensor is installed on one side of the main body. A miniature gravimeter is installed at the bottom of the main body. A temperature sensor is installed at the top of the main body. An edge computing module is installed inside the main body.

3. The lightweight oil and gas exploration system according to claim 1, characterized in that: The steps for deploying exploration units and establishing communication connections in S1 are as follows: based on the geological data of the exploration area, plan the initial distribution scheme of modular lightweight exploration units and the preset flight path of intelligent relay drones, deploy and start exploration units according to the initial distribution scheme, and establish communication connections between each exploration unit through LoRa+Mesh self-organizing network.

4. The lightweight oil and gas exploration system according to claim 3, characterized in that: In S1, the relevant raw data includes seismic wave velocity, electromagnetic impedance, gravity gradient, and geothermal field data. The edge computing module uses wavelet transform combined with Kalman filtering to eliminate noise in the acquired raw data. Specifically, the acquired raw data is first subjected to wavelet transform to decompose different frequency components, identify and filter out noise frequency band signals, and then Kalman filtering is used to dynamically estimate and optimize the remaining signal.

5. A lightweight oil and gas exploration system according to claim 4, characterized in that: In S5, the multiphysics data includes seismic wave velocity, electromagnetic impedance, gravity gradient, geothermal field, and magnetic anomaly data.

Images