Hybrid architecture based intelligent prediction system for spatial location of oil equipment

Abstract

This invention discloses a hybrid architecture-based intelligent spatial position prediction system for oil drilling equipment, comprising a human-computer interaction and control module, a binocular vision acquisition module, an image preprocessing module, a deep learning parsing module, a fusion physical modeling module, a spatial position calculation and early warning module, and an intelligent adaptive optimization module. This invention relates to the field of oil drilling rig control technology. This hybrid architecture-based intelligent spatial position prediction system for oil drilling equipment, by integrating binocular vision technology, deep learning algorithms, and physical modeling, achieves real-time prediction and dynamic adjustment of the spatial position of equipment and surrounding objects during oil drilling operations, effectively avoiding collisions between equipment and improving the safety, efficiency, and stability of oil drilling rigs.

Description

Technical Field

[0001] This invention relates to the field of oil drilling rig control technology, specifically to an intelligent prediction system for the spatial location of oil equipment based on a hybrid architecture. Background Technology

[0002] With the increasing complexity of oil drilling rig operating environments and the growing demand for intelligent systems, traditional spatial location prediction methods mainly rely on physical sensors such as lidar and infrared rangefinders. While these devices can provide real-time distance data, they suffer from problems such as system complexity, high hardware costs, difficult maintenance, and complex wiring layouts. Furthermore, oil drilling rigs typically operate in harsh environments, such as high temperature, high humidity, and strong vibration, making traditional sensors prone to failure under these conditions, impacting operational efficiency and equipment safety. To improve the safety, stability, and economic efficiency of drilling operations, there is an urgent need for an innovative method capable of high-precision, real-time, and intelligent spatial distance prediction. This patent proposes a hybrid architecture-based intelligent spatial location prediction system for oil equipment, aiming to simplify the system structure, improve prediction accuracy, and enhance stability under complex operating conditions. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides an intelligent spatial location prediction system for oil equipment based on a hybrid architecture, which solves the problem that traditional sensors are prone to failure under these operating conditions, affecting operational efficiency and equipment safety.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a hybrid architecture-based intelligent spatial location prediction system for oil equipment, comprising a human-computer interaction and control module, a binocular vision acquisition module, an image preprocessing module, a deep learning analysis module, a fusion physical modeling module, a spatial location calculation and early warning module, and an intelligent adaptive optimization module. The output of the human-computer interaction and control module is electrically connected to the input and output of the fusion physical modeling module and the intelligent adaptive optimization module, and the binocular vision acquisition module, the image preprocessing module, the deep learning analysis module, and the spatial location calculation and early warning module are electrically connected to the human-computer interaction and control module.

[0005] Preferably, the binocular vision acquisition module uses binocular cameras to acquire real-time three-dimensional image information of the oil drilling rig's operating area. The binocular vision system acquires images from different angles through synchronous cameras and generates high-precision depth information through parallax calculation to capture the relative position of the equipment and surrounding obstacles.

[0006] Preferably, the image preprocessing module processes the raw image data acquired by the binocular camera, uses Gaussian filtering for noise reduction, and Canny edge detection to enhance image edge features, ensuring that subsequent processing can extract key spatial information from the image and improve image data quality.

[0007] Preferably, the deep learning parsing module is based on preprocessed image data. This module uses a convolutional neural network to extract spatial features from the three-dimensional image, identify the relative distance and dynamic changes between the drilling equipment and surrounding objects. At the same time, this module introduces a long short-term memory network to model the motion of key components of the oil drilling rig through a multibody dynamics model. Combined with the deep learning parsing results, the motion trend of the equipment is predicted. The system uses a Kalman filter to fuse the physical model and the deep learning results to ensure the accuracy and reliability of the prediction results.

[0008] Preferably, the spatial location calculation and early warning module calculates the spatial distance between each piece of equipment on the oil drilling rig and surrounding obstacles in real time based on the results of physical modeling and deep learning analysis, and monitors the distance using a preset safety threshold. When the predicted distance is lower than the safety threshold, the system automatically issues an early warning signal to notify the operator to take necessary safety measures to avoid collisions.

[0009] Preferably, the intelligent adaptive optimization module introduces a deep reinforcement learning algorithm. The system can automatically adjust the parameter settings of the prediction model according to the real-time operating environment and equipment status of the drilling rig, ensuring that high-precision prediction can be maintained under different working conditions. In addition, the module has self-learning capabilities, and the system continuously optimizes its own algorithm during operation to adapt to new environmental changes.

[0010] Preferably, the human-machine interaction and control module displays the spatial position prediction and early warning information of the equipment to the operator through the user interface. The operator can use this module to view the operating status of the drilling equipment in real time and adjust the system parameters. This module is connected to the control system of each component of the drilling rig through a touch device, providing emergency control and operation command issuance functions for the equipment.

[0011] This invention provides an intelligent spatial location prediction system for oilfield equipment based on a hybrid architecture. Compared with existing technologies, it has the following advantages: 1. This intelligent spatial position prediction system for oil equipment based on a hybrid architecture combines binocular vision with deep learning to achieve high-precision spatial position prediction of oil drilling equipment and its surrounding environment, effectively avoiding equipment collisions and reducing the occurrence of safety accidents.

[0012] 2. This intelligent spatial location prediction system for oil equipment based on a hybrid architecture has adaptive capabilities: the introduction of deep learning and deep reinforcement learning algorithms enables the system to automatically adjust parameters and optimize prediction performance according to real-time operating conditions, ensuring that the equipment can still work efficiently in harsh environments.

[0013] 3. This intelligent spatial location prediction system for oil equipment based on a hybrid architecture simplifies hardware and reduces costs: Compared with traditional multi-sensor solutions, this invention replaces most physical ranging devices with visual sensors, simplifying hardware configuration and reducing the installation, debugging and maintenance costs of the system.

[0014] 4. This intelligent spatial positioning prediction system for oil equipment based on a hybrid architecture offers convenient operation and visual control: Through the human-machine interaction module, operators can monitor the operating status of the oil drilling rig in real time and respond quickly when necessary, ensuring the safety and efficiency of the operation process. Attached Figure Description

[0015] Figure 1 This is a system diagram of the present invention; Figure 2 This is a system control flowchart of the present invention.

[0016] The diagram shows: 1. Binocular vision acquisition module; 2. Image preprocessing module; 3. Deep learning parsing module; 4. Fusion physical modeling module; 5. Spatial position calculation and early warning module; 6. Intelligent adaptive optimization module; 7. Human-computer interaction and control module. Detailed Implementation

[0017] The technical solutions in the embodiments of the present invention have been clearly and completely described. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Please see Figure 1-2 This invention provides a technical solution: a hybrid architecture-based intelligent spatial location prediction system for oil equipment, comprising a binocular vision acquisition module 1, an image preprocessing module 2, a deep learning parsing module 3, a fusion physical modeling module 4, a spatial location calculation and early warning module 5, an intelligent adaptive optimization module 6, and a human-computer interaction and control module 7. Each module controls and transmits signals to each other to jointly achieve spatial location prediction of the path, ensuring the safe operation of oil drilling rigs.

[0019] The binocular vision acquisition module 1 is responsible for acquiring real-time three-dimensional images through binocular cameras installed in key parts of the drilling rig. The binocular cameras are arranged in different positions in the drilling rig's operating space to simultaneously acquire image data from two perspectives. The system converts the two-dimensional image data into a three-dimensional depth map through parallax calculation to provide accurate spatial distance information. The acquired image data is sent to the image preprocessing module 2 for processing through the transmission channel.

[0020] Image preprocessing module 2 is responsible for noise reduction, edge enhancement, and extraction of the raw images from binocular vision acquisition module 1. First, Gaussian excitation is used to process image noise and remove environmental noise. Then, the Canny edge detection algorithm is used to identify edge features in the image. The cropped data is then sent to deep learning parsing module 3 for further analysis. The Canny edge detection algorithm is a multi-level edge detection algorithm, and its purpose is typically to significantly reduce the image data size while preserving the original image attributes.

[0021] The deep learning parsing module 3 receives sparse image data from the image preprocessing module 2, uses a deep neural network (CNN) to extract spatial features, and identifies the relative position, distance, and changes of the equipment with surrounding dynamic objects. Simultaneously, the system introduces a long short-term memory (LSTM) network to process multi-frame image data to predict the spatial position of the equipment. The deep learning parsing module 3 calculates the future positions of each piece of equipment on the drilling rig using a deep learning model, and transmits the fused parsing results to the fused physical modeling module 4.

[0022] The fusion physical modeling module 4 models the motion of the equipment by constructing a multibody dynamics model of key components of the oil drilling rig. This model is based on the kinematic parameters of the drilling rig, the geometric structure of the equipment, and the operating environment to simulate the future motion of the equipment. The fusion physical modeling module 4 fuses the calculation results of the physical model with the deep learning parsing data from the deep learning parsing module 3 through a Kalman filter to ensure the accuracy of the prediction. The fused data is then transmitted to the spatial position calculation and early warning module 5.

[0023] The spatial position calculation and early warning module 5 calculates the spatial distance between each piece of drilling equipment and surrounding objects in real time based on the fusion data transmitted by the fusion physical modeling module 4. The spatial position calculation and early warning module 5 monitors the operation of the equipment by setting a safe distance threshold. When the distance between the equipment and the obstacle is close to or lower than the preset threshold, the system automatically generates an early warning signal and sends it to the human-machine interaction and control module 7 for display. The module also issues an alarm to the operator.

[0024] Meanwhile, the spatial location calculation and early warning module 5 transmits the safe distance information between devices to the intelligent adaptive optimization module 6 so that the system can make optimization adjustments based on real-time conditions.

[0025] The intelligent adaptive optimization module 6 introduces the deep reinforcement learning (DRL) algorithm to achieve adaptive optimization of the system. When the drilling rig's operating environment or equipment status changes, the intelligent adaptive optimization module 6 automatically adjusts the system's parameters and model settings based on real-time environmental data and equipment status to ensure the position and stability of spatial location prediction. The intelligent adaptive optimization module 6 also has continuous self-learning capabilities and can optimize its own algorithm through accumulated operating data to improve the system's ability to cope with complex working conditions.

[0026] The human-machine interaction and control module 7 provides operators with a visual display and control interface for the equipment's operating status. The human-machine interaction and control module 7 displays the spatial position prediction results and early warning information of the equipment to the operator through a touch screen or other display device. When the spatial position calculation and early warning module 5 issues an early warning signal, the operator can view the necessary real-time distance between the equipment and the operating status through the human-machine interaction and control module 7 and intervene in the operation. The human-machine interaction and control module 7 also allows the operator to input operating commands to manually control the movement of the equipment, ensuring the flexibility of the system under special working conditions.

[0027] The human-computer interaction and control module 7, along with modules such as image preprocessing module 2, deep learning parsing module 3, and fusion physical modeling module 4, are connected through a system interconnection module to ensure signal transmission and control coordination between the modules.

[0028] Furthermore, any content not described in detail in this specification is existing technology known to those skilled in the art.

[0029] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0030] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A hybrid architecture based intelligent prediction system for spatial location of oil equipment comprising of human machine interface and control module (7) characterized by: It also includes a binocular vision acquisition module (1), an image preprocessing module (2), a deep learning analysis module (3), a fusion physical modeling module (4), a spatial position calculation and early warning module (5), and an intelligent adaptive optimization module (6). The output end of the human-computer interaction and control module (7) is electrically connected to the input and output ends of the fusion physical modeling module (4) and the intelligent adaptive optimization module (6), and the binocular vision acquisition module (1), the image preprocessing module (2), the deep learning analysis module (3), and the spatial position calculation and early warning module (5) are electrically connected to the human-computer interaction and control module (7).

2. The hybrid architecture based intelligent spatial location prediction system for oil and gas equipment as claimed in claim 1, wherein: The binocular vision acquisition module (1) uses a binocular camera to acquire real-time three-dimensional image information of the oil drilling rig's operating area.

3. The intelligent spatial location prediction system for oil equipment based on a hybrid architecture according to claim 1, characterized in that: The image preprocessing module (2) is used to process the raw image data acquired by the binocular camera.

4. The intelligent spatial location prediction system for oil equipment based on a hybrid architecture according to claim 1, characterized in that: The deep learning parsing module (3) extracts spatial features from the three-dimensional image using a convolutional neural network (CNN) based on the preprocessed image data.

5. The intelligent spatial location prediction system for oil equipment based on a hybrid architecture according to claim 1, characterized in that: The fusion physical modeling module (4) models the motion of key components of the oil drilling rig through a multibody dynamics model.

6. The intelligent spatial location prediction system for oil equipment based on a hybrid architecture according to claim 1, characterized in that: The spatial location calculation and early warning module (5) calculates the spatial distance between each piece of equipment of the oil drilling rig and surrounding obstacles in real time based on the results of the physical model and deep learning analysis.

7. The intelligent spatial location prediction system for oil equipment based on a hybrid architecture according to claim 1, characterized in that: The intelligent adaptive optimization module (6) introduces the deep reinforcement learning (DRL) algorithm, which can automatically adjust the parameter settings of the prediction model according to the real-time working environment and equipment status of the drilling rig, ensuring that high-precision prediction can be maintained under different working conditions, and has self-learning ability.

8. The intelligent spatial location prediction system for oil equipment based on a hybrid architecture according to claim 1, characterized in that: The human-computer interaction and control module (7) displays the spatial location prediction and early warning information of the equipment to the operator through the user interface.

Images