Intelligent monitoring and full-process simulation optimization system for the central control system of offshore oil platforms

Abstract

The present invention belongs to the technical field of central control platforms, and specifically relates to an intelligent monitoring and full-process simulation optimization system for a central control system of an offshore oil platform, including an offshore oil platform central control system, wherein the offshore oil platform central control system is internally provided with an intelligent monitoring platform and a full-process simulation optimization platform, wherein the intelligent monitoring platform is internally provided with an abnormal history and analysis module, a fault prediction and optimization module, and an unmanned inspection integration module; wherein the full-process simulation optimization platform is internally provided with a full-process simulation optimization module and an intelligent instrument ledger module, and further comprises drones and engine rooms arranged in high-risk areas corresponding to offshore oil platforms, and both the drones and engine rooms are designed with corrosion-resistant fuselages. The present invention has a multi-level emergency response mechanism and the ability to cope with extreme working conditions, adopts full-process simulation optimization, intelligent inspection, and predictive maintenance, and has both a multimodal data fusion architecture and an open platform expansion.

Description

Technical Field

[0001] The present invention belongs to the technical field of central control platforms, and in particular relates to an intelligent monitoring and full-process simulation optimization system for a central control system of an offshore oil platform. Background Art

[0002] Offshore oil production is characterized by high risk, high cost, and strong environmental sensitivity. The volatile marine climate (such as typhoons and wave impacts) and high risk of equipment corrosion pose a threat to the safe operation of the platform. Traditional central control systems rely on manual experience-based decision-making, and emergency response lags, which can easily lead to major accidents such as leaks and fires. Therefore, the industrial digital offshore oil central control platform, leveraging the development of the Internet of Things, big data, and AI, optimizes the control of oil platforms.

[0003] Problems with existing technologies:

[0004] Safety and efficiency bottlenecks: Manual inspections are highly risky, as evidenced by the high accident rate associated with traditional high-altitude inspections; delayed emergency response, as evidenced by the 5-8 minutes it takes to manually trigger the platform shutdown system; and inefficient process control, as evidenced by the lack of advance optimization of process parameters and a low control stability rate.

[0005] Data silos and system fragmentation: Multi-source data is fragmented, with SCADA, video surveillance, and equipment logs belonging to independent systems without feature fusion, resulting in low data analysis efficiency.

[0006] Poor adaptability to extreme working conditions and complex environments: The risk of uncontrolled typhoon impacts is manifested as a significant increase in the failure rate under level 7 wind conditions. In addition, regarding the problem of equipment failure caused by salt spray corrosion, the system platform does not classify the failure causes of the above extreme situations, and is therefore unable to identify high-corrosion areas in advance based on experience, and is unable to improve the coverage of anti-corrosion measures and typhoon prevention measures. Summary of the Invention

[0007] The purpose of this invention is to provide an intelligent monitoring and full-process simulation optimization system for the central control system of an offshore oil platform, which can have a multi-level emergency response mechanism and the ability to cope with extreme working conditions, adopt full-process simulation optimization, intelligent inspection and predictive maintenance, and have both multimodal data fusion architecture and open platform expansion.

[0008] The technical solutions adopted by the present invention are as follows:

[0009] An intelligent monitoring and full-process simulation optimization system for an offshore oil platform central control system includes an offshore oil platform central control system, wherein the offshore oil platform central control system is internally provided with an intelligent monitoring platform and a full-process simulation optimization platform;

[0010] The internal configuration of the intelligent monitoring platform includes:

[0011] The abnormal history and analysis module comprehensively analyzes the abnormal causes through multi-dimensional statistics. It has an abnormal cause classification module and an abnormal result analysis module.

[0012] Fault prediction and optimization module, which performs real-time simulation and fault prediction of platform equipment operating status and process flow;

[0013] The unmanned inspection integration module cooperates with drones and cabins distributed in corresponding areas to automatically inspect high-risk areas. It is equipped with a drone control intervention module, a navigation interface module, an inspection video module, and a hovering inspection function module.

[0014] The internal configuration of the full-process simulation optimization platform includes:

[0015] The full-process simulation optimization module integrates full-process semi-physical simulation technology and heterogeneous information processing methods, supports digital twin modeling, and performs real-time simulation of platform equipment operating status and process flow;

[0016] The smart instrument ledger module systematically manages the equipment ledger and classifies and counts smart instruments based on dimensions such as device region, manufacturer, system, and device level.

[0017] It also includes drones and cabins deployed in high-risk areas corresponding to offshore oil platforms, and both drones and cabins adopt an anti-corrosion fuselage design.

[0018] The intelligent monitoring platform is also equipped with a device configuration monitoring module, an early warning statistics and suggestion module, and an abnormal alarm module;

[0019] The equipment configuration monitoring module is used to collect status signals from smart instruments and alarm signals from the platform environment. The equipment configuration monitoring module is internally equipped with an equipment status acquisition module and a multi-dimensional data integration module. The equipment status acquisition module divides the status signals of smart instruments into five levels according to severity: no communication, fault, abnormality, maintenance, and suggestion. The multi-dimensional data integration module integrates various abnormal environment alarm signals from the oil platform.

[0020] The abnormal cause classification module is used to integrate and analyze abnormal device status signals and classify factors that cause device abnormalities; the abnormal result analysis module provides corresponding countermeasures based on the type of abnormal cause.

[0021] The early warning statistics and suggestion module is used to display the fault diagnosis information and historical events of the central control system's alarm equipment today in real time, and provide guidance and suggestions.

[0022] The abnormal alarm module is used to issue an alarm when abnormal data is detected.

[0023] The drone control intervention module conducts human-machine collaborative inspection and data analysis through the generated control interface;

[0024] The navigation interface module plans three-dimensional paths and generates drone navigation routes;

[0025] The inspection video module is used to track and record the inspection process in real time;

[0026] The hovering inspection function module develops a hovering inspection function to ensure that the drone is in a relatively static state for inspection when the oil platform is shaking.

[0027] The full-process simulation optimization platform is also provided with a system architecture overview module and a system evaluation module.

[0028] The system architecture overview module systematically introduces the system scale of the central control system, displays the overall architecture of the central control system, and dynamically displays the real-time online status of the operation station, the communication status of the central control system controller, and the RS communication status between the central control and third-party equipment.

[0029] The system evaluation module provides an overall score for the central control system based on its current system status, fault information and other static data as well as real-time status data, and proposes feasible optimization suggestions.

[0030] The full-process simulation optimization module is internally provided with a semi-physical simulation module, a heterogeneous information processing module and a digital twin modeling module;

[0031] The semi-physical simulation module is used to implement the full-process semi-physical simulation technology; the heterogeneous information processing module is used to process and fuse heterogeneous information; and the digital twin modeling module is used to perform digital twin modeling and synchronization.

[0032] The technical effects achieved by the present invention are:

[0033] (1) Upgrading security assurance and risk control capabilities

[0034] Multi-level emergency response mechanism: Integrates three independent systems: process control, emergency shutdown, and fire and gas detection, achieving safety redundancy through logical linkage; the equipment weight grading module dynamically calculates the risk values ​​of A / B / C level instruments and triggers graded emergency plans.

[0035] Ability to cope with extreme working conditions: The digital twin module is combined with the platform sway compensation algorithm to simulate the impact of tilt angle on the equipment in typhoon mode to ensure control stability.

[0036] (2) Efficiency Improvement and Intelligent Innovation

[0037] Full-process simulation optimization: Semi-physical simulation technology uses a dynamic model library to pre-optimize process parameters and improve control stability;

[0038] Intelligent inspection efficiency increased: The drone's hovering inspection module uses a wind-resistant positioning algorithm, combined with 3D real-scene modeling navigation, to increase inspection efficiency in high-risk areas by three times;

[0039] Predictive maintenance innovation: The LSTM time series prediction model provides early warning of Class A equipment failures 4-8 hours in advance, shortening maintenance response time to 40 minutes per 1,000 reports.

[0040] (3) System integration and expansion capabilities

[0041] Multimodal data fusion architecture: Tensor decomposition technology integrates heterogeneous data such as SCADA signals, video streams, and equipment logs, achieving a feature extraction accuracy of 92%;

[0042] Open platform expansion: Data interoperability with third-party systems is achieved through the OPC UA protocol, and plug-in upgrades of functional modules are supported. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a diagram showing the composition of the central control system provided by an embodiment of the present invention;

[0044] Figure 2 This is a schematic diagram of the distribution of UAV cabins in an offshore oil platform and a corresponding area provided by an embodiment of the present invention.

[0045] In the accompanying drawings, the components represented by the reference numerals are as follows:

[0046] 1. Offshore oil platform central control system; 2. Intelligent monitoring platform; 21. Equipment configuration monitoring module; 211. Equipment status acquisition module; 212. Multi-dimensional data integration module; 22. Abnormal history and analysis module; 221. Abnormal cause classification module; 222. Abnormal result analysis module; 23. Early warning statistics and suggestion module; 24. Fault prediction and optimization module; 25. Abnormal alarm module; 26. Unmanned inspection integration module; 261. UAV control intervention module; 262. Navigation interface module; 263. Inspection video module; 264. Hovering inspection function module; 3. Full-process simulation and optimization platform; 31. Full-process simulation and optimization module; 311. Semi-physical simulation module; 312. Heterogeneous information processing module; 313. Digital twin modeling module; 32. Smart instrument ledger module; 33. System architecture overview module; 34. System evaluation module. DETAILED DESCRIPTION

[0047] In order to make the purpose and advantages of the present invention more clearly understood, the present invention is described in detail below with reference to the following examples. It should be understood that the following text is only used to describe one or more specific embodiments of the present invention and does not strictly limit the scope of protection of the present invention.

[0048] like Figure 1-Figure 2 As shown, the offshore oil platform central control system intelligent monitoring and full-process simulation optimization system, the offshore oil platform central control system 1, the offshore oil platform central control system 1 is internally equipped with an intelligent monitoring platform 2 and a full-process simulation optimization platform 3.

[0049] The intelligent monitoring platform 2 is internally provided with a device configuration monitoring module 21, an abnormality history and analysis module 22, an early warning statistics and suggestion module 23, a fault prediction and optimization module 24, an abnormality alarm module 25, and an unmanned inspection integration module 26;

[0050] The full-process simulation optimization platform 3 is internally provided with a full-process simulation optimization module 31 , an intelligent instrument ledger module 32 , a system architecture overview module 33 and a system evaluation module 34 .

[0051] The equipment configuration monitoring module 21 is used to collect status signals of smart instruments and alarm signals of the platform environment, which includes an equipment status acquisition module 211 and a multi-dimensional data integration module 212. The equipment status acquisition module 211 divides the status signals of smart instruments into five levels according to severity: no communication, fault, abnormality, maintenance, and suggestion. The multi-dimensional data integration module 212 integrates various abnormal environment alarm signals of the oil platform, including shaking caused by typhoons / waves, high salt fog environment, high humidity environment, etc. The above signal acquisition technologies are all existing mature means and will not be elaborated here.

[0052] The Abnormal History and Analysis Module 22 uses multi-dimensional statistics to comprehensively analyze the causes of abnormalities, including device type and manufacturer, environmental factors, and operational factors, to gain a deep understanding of the root cause of the alarm;

[0053] It is internally provided with an abnormality cause classification module 221 and an abnormality result analysis module 222. The abnormality cause classification module 221 is used to integrate and analyze the abnormality signals of the equipment status, and classify the factors causing the equipment abnormality by combining the environmental information when the abnormality occurs, the equipment manufacturer information, the oil platform shaking information and the operation signal; the abnormality result analysis module 222 provides corresponding countermeasures according to the type of abnormality cause;

[0054] The specific process is as follows:

[0055] Step 1: Multimodal data collection and feature extraction

[0056] Please refer to Table 1 below:

[0057] Table 1

[0058] Data Type Collection method Key Parameter Examples Preprocessing technology Device status signal Smart meter + edge computing node Vibration spectrum>15kHz, temperature gradient ΔT>8℃ / s Wavelet Noise Reduction Environmental Information Weather Station + Corrosion Sensor <![CDATA[Salt spray concentration > 5mg / m 3 , humidity > 90%RH]]> Kalman filter calibration Platform motion parameters Inertial Navigation System (INS) Roll angle>2°, heave displacement>1.5m Fourier transform frequency domain analysis Equipment manufacturer data Central database API call Sealing level IP68, MTBF>100,000 hours Ontological reasoning

[0059] Example: When a pump on the Qikou 17-2 platform vibrates abnormally, the salt spray concentration is collected simultaneously at 8.2 mg / m 3 (exceeding the threshold by 53%), and the platform roll angle was 3.1° (exceeding the design limit).

[0060] Step 2: Feature Fusion and Association Modeling

[0061] Tensor fusion engine: Constructs a four-dimensional feature tensor, using Tucker decomposition to extract core features;

[0062] Causal diagram construction: Using a causal discovery algorithm (PC algorithm) based on historical data, a fault tree (FTA) is constructed to define the logical relationships between abnormal events. Based on the above examples, the insulation resistance can be inferred from the salt spray concentration, and the bearing displacement can be inferred from the platform roll angle.

[0063] Step 3: Dynamic Clustering and Inducement Classification

[0064] Bayesian network real-time update (P(F j |E)):

[0065] Evidence E: Salt spray concentration 8.2 mg / m 3 +Siemens motor sealing level IP66;

[0066] Hypothesis F: F1: salt spray corrosion causes short circuit (prior probability P(F1)=0.32), F2: platform shaking causes mechanical fatigue (P(F2)=0.41);

[0067] Posterior calculation: P(F1|E)=0.79 → Determined to be an environmental trigger.

[0068] Based on the above content: Give examples of classification rules for abnormal types:

[0069] Environmental inducements: Salt spray concentration> 5mg / m 3 And the equipment sealing level <IP65;

[0070] Mechanical inducement: vibration energy>0.1g 2 / Hz and bearing life < 80% of design value;

[0071] Operational inducement: Platform tilt angle > 2° and compensation mechanism not triggered;

[0072] By integrating the above-mentioned equipment abnormality causes, the corresponding equipment can be controlled to take countermeasures when the abnormal cause occurs again, for example:

[0073] Scenario 1: Instrument communication interruption: Traditional solutions rely on manual line troubleshooting, which is time-consuming. This solution attributes the problem to salt spray corrosion within 3 minutes and activates redundant channels, improving recovery time.

[0074] Scenario 2: Pump body vibration exceeds the standard: The traditional solution involves shutting down the machine for disassembly and inspection, which results in economic losses. This solution dynamically determines if the vibration causes shaft offset and automatically compensates with counterweights, avoiding unplanned production stoppages.

[0075] Scenario 3: Valve Response Delay: The traditional solution involves replacing the entire valve, which is costly. This solution accurately identifies salt crystal accumulation on the seal ring and performs targeted flushing, reducing maintenance costs.

[0076] By identifying the causes of frequent alarms in devices, the operation of smart meters can be effectively managed and optimized, improving system stability and reliability, while also improving maintenance and operation efficiency.

[0077] The early warning statistics and suggestion module 23 monitors the alarms of the equipment and displays the alarm equipment of a single day in the form of a chart. It also displays the fault diagnosis information and historical events of the alarm equipment of the central control system in real time. It provides feasible expert guidance and suggestions on the alarm information of the equipment, and has the functions of suppressing equipment alarms and managing planned disposal, so that the equipment alarm forms a closed-loop management. The technology of displaying alarm information in a graphical way and providing guidance and suggestions is an existing mature method and will not be elaborated on here.

[0078] Fault prediction and optimization module 24, which performs real-time simulation and fault prediction of platform equipment operating status and process flow;

[0079] The specific process is as follows:

[0080] Step 1: Real-time perception of global data

[0081] Please refer to Table 2 below:

[0082] Table 2

[0083] Data Type Collection method Typical parameters Processing Technology Device Status Smart meter + edge computing node Vibration spectrum, temperature gradient, current harmonics Millisecond sampling + wavelet noise reduction Process SCADA+Multiphase Flow Sensor Oil / gas / water ratio, pipeline pressure drop, separation efficiency Multi-source data spatiotemporal alignment technology marine environment Weather station + platform tilt sensor Wind speed, salt spray concentration, platform roll angle Enhanced design against electromagnetic interference Historical Knowledge Base 10-year fault record + manufacturer's maintenance manual Failure mode library, maintenance solution library Ontology push modeling

[0084] Example scenario: When a typhoon passes, the system simultaneously captures a sudden increase of 35% in compressor bearing vibration, a platform roll angle exceeding the design value by 2.8°, and a separator liquid level fluctuation exceeding ±15%.

[0085] Step 2: Real-time simulation of digital twins

[0086] Multi-physics dynamic modeling: Establish a three-dimensional virtual image of the platform equipment, including the mapping of mechanical structure stress distribution (such as vortex-induced vibration of pipeline elbows), fluid dynamics models (such as oil-gas-water separator flow), and electrical system status (such as inverter heat loss);

[0087] Working condition simulation engine: Normal working condition: simulates theoretical operating status under current parameters; extreme working condition: injects environmental disturbance parameters such as typhoon and high salt fog; fault simulation: automatically triggers 200+ failure modes (such as bearing fracture and valve sticking);

[0088] Bohai Sea test case: In the simulation of "level 7 wind + salt spray 8mg / m 3 "Under the working condition, the risk of fire pump seal failure was warned 12 minutes in advance, and the similarity between the actual fault characteristics and the simulation reached 91%.

[0089] Step 3: Fault prediction and root cause analysis

[0090] Multi-model collaborative diagnosis: Using a time series prediction model, it outputs a remaining life estimate (error <8%) for gradual faults (such as equipment wear); using an anomaly detection model, it outputs fault location for sudden faults (such as circuit short circuit); and using a causal reasoning engine, it outputs a root cause contribution ranking for complex faults (such as shaking + corrosion).

[0091] Dynamic risk assessment matrix: Recommended measures are given based on the equipment ID information, combined with the risk level and main causes. For example: Equipment ID number P-203A is at a critical level, the main causes are salt spray corrosion and vibration fatigue, and it is recommended to shut down the equipment within 2 hours for corrosion prevention.

[0092] Step 4: Autonomous decision-making and closed-loop control

[0093] Gradual response mechanism: Level 1 (low risk): automatically adjust PID parameters for compensation; Level 2 (medium risk): activate equipment redundancy system; Level 3 (high risk): trigger ESD emergency shutdown;

[0094] Knowledge self-evolution process: Compare the actual fault data with the simulation results, analyze the two, and when the error is determined to be greater than 10%, modify the model parameters. When the error is determined to be less than 10%, store it in the successful case library.

[0095] Based on the above content: Give typical applications:

[0096] Scenario 1: Compressor vibration exceeded standards during a typhoon: The traditional solution involved manual load reduction, resulting in a 15% production reduction. This solution, by predicting bearing displacement risks and implementing automatic balancing, avoided production reductions and mitigated losses.

[0097] Scenario 2: Hydrate blockage in submarine pipelines: The traditional solution involves shutting down production and injecting methanol, which is time-consuming. This solution provides a 4-hour advance warning and adjusts temperature and pressure parameters, achieving zero production downtime.

[0098] Scenario 3: Generator winding overheating: The traditional solution uses passive shutdown for maintenance, resulting in a 15% production reduction. This solution identifies cooling fan failures and remotely switches to a backup, reducing maintenance costs.

[0099] Statistical verification shows that the false alarm rate has dropped to 5.2% (industry average 23%), the incidence of major accidents has dropped by 82%, and annual operation and maintenance costs have been reduced.

[0100] The abnormal alarm module 25 is used to alarm when abnormal data is monitored or when the fault prediction and optimization module 24 predicts that a fault is about to occur. The above alarm technologies are all existing mature means and will not be described in detail here.

[0101] The unmanned inspection integration module 26 cooperates with drones and cabins distributed in the corresponding areas to automatically inspect high-risk areas and reduce the risk of manual exposure;

[0102] The specific process is as follows:

[0103] Step 1: Exception triggering and task scheduling;

[0104] 1.1) Anomaly detection and priority determination:

[0105] The central control system monitors sensor data (temperature, vibration, pressure, etc.) of equipment in high-risk areas (such as pressure vessels and oil pipelines) in real time. When it detects that a parameter exceeds a safety threshold or the AI ​​model identifies an abnormal pattern, it automatically triggers an inspection request.

[0106] Generate drone mission instructions of different priorities based on the anomaly type (such as leakage, overheating) and risk level (refer to the A / B / C level equipment weights in historical conversations).

[0107] 1.2) UAV task allocation: Optimize multi-UAV scheduling based on the improved Hungarian algorithm.

[0108] Step 2: Using the navigation interface module 262 to perform three-dimensional path planning and navigation;

[0109] 2.1) Real scene modeling and navigation line generation:

[0110] Real-scene modeling call: The system calls a pre-built 3D real-scene model (generated by laser scanning or ContextCapture), combining platform structure, equipment layout, and safety restricted area information;

[0111] Dynamic Path Generation: Automatically plans the optimal inspection path based on the target equipment location and real-time environmental data (such as wind speed and obstacles), avoiding dangerous structures such as pipelines and towers.

[0112] 2.2) Coordinate mapping and flight control:

[0113] One-click dispatch and takeoff: After the staff confirms the mission through the control interface, the drone automatically takes off from the explosion-proof hangar and flies to the target area along the preset path;

[0114] Adaptive flight adjustment: The drone is equipped with RTK high-precision positioning and vision sensors, which compare the coordinates of the 3D model in real time and dynamically correct the flight trajectory (such as position compensation when the platform shakes).

[0115] Step 3: Use the drone control intervention module 261 to conduct human-machine collaborative inspection and data analysis;

[0116] Control interface generation: Build an AR interactive interface based on Unity3D, rendering the drone's posture and device status parameters in real time. Workers can manually adjust the drone's posture, lens focal length, or temporarily add detection points;

[0117] Abnormal data collection and processing: Multispectral cameras capture images, detect equipment defects using YOLOv5, and output defect locations and categories (such as rust and cracks). Thermal imaging data is then integrated with mechanism models for analysis.

[0118] The inspection video module 263 is used to track and record the inspection process in real time to provide video evidence for other re-inspections. This tracking video technology is an existing mature means and will not be described in detail here.

[0119] Step 4: Utilize the hovering inspection function module 264 to develop a hovering inspection function, which is used to ensure that the drone is in a relatively stationary state for inspection when the oil platform is shaking.

[0120] 4.1) Dynamic compensation of platform shaking:

[0121] The platform-mounted differential GPS base station and inertial navigation system (INS) are used to calculate the displacement and angular deviation caused by platform shaking in real time, thus establishing a dynamic reference coordinate system.

[0122] The drone receives the compensation signal and automatically adjusts its flight attitude to offset the relative displacement error caused by the platform shaking and maintain a fixed relative position with the platform equipment.

[0123] 4.2) Wind-resistant hovering position optimization:

[0124] Multi-source environmental perception: The drone is equipped with an ultrasonic anemometer, barometer, and visual sensors to monitor the wind distribution in the inspection area in real time and identify instantaneous wind speed changes. This data is transmitted back to the central control system via a 5G private network, and combined with the real-time posture data of the platform's swaying to generate a 3D wind field model.

[0125] Adaptive hovering adjustment: Based on the wind field model, the drone's flight control system dynamically adjusts the blade speed and fuselage inclination angle to combat crosswind interference. Through the edge computing module, it makes quick decisions and prioritizes hovering in low wind speed areas on the leeward side of the platform structure or in equipment-intensive areas.

[0126] 4.3) Optimal hovering point calculation

[0127] 3D spatial modeling: Using the digital twin platform's high-precision 3D model (such as point cloud data generated by laser scanning), we can annotate the platform structure's obstruction of airflow. Combined with historical inspection data, we can create turbulence intensity heat maps for each area under different wind direction conditions.

[0128] Real-time path optimization: The central control system dynamically plans hovering points based on current wind speed, direction, and platform sway parameters using the A* algorithm to avoid high-turbulence areas. The recommended hovering position is intuitively displayed through an augmented reality (AR) interface to assist operators in decision-making.

[0129] 4.4) Human-machine collaborative control

[0130] Intelligent control interface: The central control room provides a 3D visual operation interface, showing the drone's real-time viewing angle, equipment status, and hovering stability parameters. It also supports one-click activation of "anti-sway mode" to automatically lock onto preset inspection targets (such as valves and instrument panels).

[0131] Semi-autonomous inspection mode: When the drone is in hovering mode, it uses the onboard infrared camera and lidar to scan the equipment from multiple angles. Abnormal data automatically triggers high-definition photography. The operator can manually fine-tune the hovering position to focus on re-inspecting high-risk equipment.

[0132] 4.5) Emergency risk avoidance mechanism

[0133] Safety threshold warning: When the wind speed exceeds level 6 or the platform tilt angle exceeds 3°, the system automatically switches to emergency mode and controls the drone to return to the platform anchor point; in the event of sudden communication interruption, the drone activates the offline navigation algorithm and returns along the preset safe path.

[0134] 4.4) Closed-loop data management

[0135] Inspection data fusion: Images, videos, and sensor data collected by drones are automatically linked with equipment records and historical fault databases through the digital twin platform to generate a comprehensive diagnostic report. Abnormal data (such as instrument reading deviations) are pushed to the maintenance work order system in real time, forming a "monitoring-analysis-disposal" closed loop.

[0136] According to the above content: based on the abnormal information of the corresponding equipment status in the high-risk area, the drone inspection function is triggered. Through the generated drone intervention control interface, along the navigation route generated by the three-dimensional real-scene modeling in the previous article, the staff can control the drone to inspect the high-risk area, which can significantly improve the inspection efficiency and shorten the inspection time of the high-risk area to 30% of the manual inspection time. At the same time, the multi-spectral + YOLOv5 combined detection is used to improve the defect recognition rate. At the same time, through the wind-resistant algorithm (such as the closed-loop control system based on PID) and the anti-corrosion fuselage design, the drone can be used in salt spray concentrations > 5mg / m 3, hover stably under wind speed of level 7, breaking through the limitations of manual inspection environment.

[0137] The full-process simulation optimization module 31 integrates the full-process semi-physical simulation technology and heterogeneous information processing methods, supports digital twin modeling, and simulates the operating status of platform equipment and process flow in real time. It is internally equipped with a semi-physical simulation module 311, a heterogeneous information processing module 312, and a digital twin modeling module 313;

[0138] The specific process is as follows:

[0139] The first step is to use the semi-physical simulation module 311 to achieve hardware-in-the-loop (HIL) simulation and real-time data coupling;

[0140] Dynamic access to physical devices: Physical signals from platform sensors (such as pressure, temperature, and flow sensors) and controllers are accessed in real time through I / O cabinets, establishing a closed-loop interaction between hardware and virtual models. For example, a 4-20mA analog signal from a field instrument is converted into a digital signal through the I / O cabinet and transmitted to the central control workstation.

[0141] Real-time response of virtual models: Mechanism models such as fluid mechanics and equipment dynamics (e.g., pipeline pressure dynamic models) receive physical signal inputs and simulate the response behavior of equipment under real working conditions, such as valve opening adjustment in typhoon scenarios.

[0142] Step 2: Using the heterogeneous information processing module 312 to perform heterogeneous data fusion and intelligent analysis;

[0143] Multi-source data integration: Synchronously processes structured data (SCADA system signals) and unstructured data (camera videos, maintenance report text), using tensor decomposition technology to extract key features and eliminate data silos. For example, environmental parameters such as salt spray concentration and platform tilt angle can be correlated with equipment vibration spectrum for analysis.

[0144] Federated learning collaborative optimization: Share encrypted data features (non-raw data) across platforms and jointly train fault prediction models to ensure data privacy while improving model generalization capabilities.

[0145] Step 3: Use the digital twin modeling module 313 to perform digital twin modeling and three-dimensional visualization;

[0146] High-precision reality modeling: Generate a 3D mesh model of the oil platform based on laser scanning and ContextCapture technology, mapping the spatial coordinates of the physical equipment;

[0147] Dynamic data binding and synchronization: Real-time data (such as valve opening and temperature readings) is bound to the 3D model via the OPC UA protocol, enabling instant updates of device status in the virtual scene. For example, the central control room screen simultaneously displays a simulated animation of crude oil flow in the pipeline and the actual pressure curve.

[0148] Step 4: Real-time simulation and decision support applications

[0149] Dynamic process simulation: simulate the entire crude oil processing process (such as separation, pressurization, and external transmission) and preview the impact of different operating conditions (such as production fluctuations and equipment start-up and shutdown) on system stability;

[0150] Emergency scenario simulation: For extreme events such as typhoons and leaks, safety interlock logic (such as ESD emergency shutdown) is triggered in the digital twin to automatically generate response plans;

[0151] Maintenance decision optimization: Based on real-time equipment performance deviations (such as decreased pump efficiency), targeted maintenance work orders are pushed and the system recovery effect after maintenance is simulated.

[0152] Smart Instrument Ledger Module 32 systematically manages equipment records and categorizes and compiles statistics on smart instruments by device region, manufacturer, system, and device level, allowing users to quickly query and compile statistics on equipment information, improving management efficiency.

[0153] The specific process is as follows:

[0154] Step 1: Equipment Weight Grading Operation Process

[0155] 1.1) Hierarchical initialization

[0156] Data input: Import basic instrument data (installation location, manufacturer model, and process unit), and mark the instrument's control nodes in the process (such as reactor pressure sensor and crude oil outlet flow meter) based on the process flow chart.

[0157] Risk quantification: Automatically calculate the initial risk value based on the historical fault database to collect statistics on the downtime, economic losses, and safety incident frequency caused by instrument failures;

[0158] Manual calibration: Process engineers confirm the default level (directly marked as Class A) of critical instruments (such as safety interlock valves).

[0159] 1.2) Dynamic Weight Adjustment

[0160] Real-time monitoring: The system links SCADA real-time data. When an instrument triggers an alarm frequently (e.g., exceeding the limit three times within 24 hours), the weight level is automatically increased (e.g., from Class B to Class A).

[0161] External factors integration:

[0162] Environmental data: In high salt fog weather, corrosion-prone instruments (such as liquid level gauges exposed on the deck) are automatically marked as Class B or above; Platform status: When the platform inclination exceeds the threshold, the weight of vibration-sensitive instruments (such as gas chromatographs) is temporarily increased.

[0163] 1.3) Classified application scenarios

[0164] Class A instrument: Real-time monitoring frequency is increased to seconds, and in the event of a fault, an audible and visual alarm is automatically triggered and synchronized to the central control screen; maintenance priority is the highest, and the maintenance closed loop must be completed within 48 hours;

[0165] Class B instrument: patrols data hourly and pushes work orders to the team's mobile terminal when an anomaly occurs;

[0166] Class C instruments: Covered by daily inspections, and faults are included in the planned maintenance queue.

[0167] Step 2: Diagnostic report management operation process

[0168] 2.1) Automated report generation

[0169] Data source integration: Valve diagnostic data (such as travel time and sealing performance) is automatically collected through the HART protocol; manual inspection reports (such as corrosion thickness reports) are uploaded to the system by scanning the code;

[0170] Intelligent classification: The report is linked to the ledger by instrument ID, and the fault type (such as "valve stem stuck" and "sensor drift") is automatically marked.

[0171] 2.2) Statistical Analysis Engine

[0172] High-frequency fault location: Automatically calculates the top three causes of failure for similar instruments (e.g., salt spray corrosion accounts for 60% of the failures of a certain manufacturer's pressure gauges) and generates a monthly fault heat map.

[0173] Maintenance decision support: When a Class B or higher instrument experiences repeated failures, the system will provide replacement recommendations (e.g., "If the same valve is diagnosed abnormal five times within three months, it is recommended that it be scrapped").

[0174] 2.3) Closed-loop management mechanism

[0175] Task distribution: The diagnostic report triggers a maintenance work order, which is automatically assigned to the responsible team and associated with spare parts inventory information (e.g., replacement valve inventory number V2037);

[0176] Electronic signature: After the maintenance is completed, the on-site scanner forms the QR code to close the work order, and the report is automatically archived.

[0177] Step 3: System linkage and value verification

[0178] Typhoon mode emergency response: When the meteorological system issues a typhoon warning, the A-level instrument status is automatically displayed at the top. If an abnormality is detected (such as a delayed response of the emergency shut-off valve), the backup power supply is forced to start;

[0179] Improved management efficiency: After the application of a certain platform, the response time for Class A instrument failures was shortened to 2 hours (originally 8 hours), and the efficiency of diagnostic report analysis increased by 70%.

[0180] Based on the above content: According to the role of smart instruments in the process, the corresponding equipment weights are decomposed and divided into Class A, Class B, and Class C; Class A refers to smart instruments that cause huge losses and cause partial equipment or the entire plant to stop; Class B refers to smart instruments that cause large losses and cause production reductions in some equipment; Class C refers to smart instruments with relatively small losses; instrument valve diagnostic reports are statistically managed through file management; in addition, through manual review and comparison, it is found that the accuracy of weight classification is significantly improved, the efficiency of diagnostic report analysis is significantly improved, and through LSTM time series prediction enhancement, early warning of Class A equipment failures can be achieved.

[0181] System Architecture Overview Module 33 systematically introduces the central control system's system scale, operating time, alarm status, controller load and trend statistics, power supply voltage monitoring, and temperature monitoring. It also systematically displays the overall architecture of the central control system and dynamically displays the real-time online status of the operation station, the communication status of the central control system controller, and the communication status between the central control system and third-party devices.

[0182] Through intuitive data display and visual charts, it helps instrumentation and control personnel to comprehensively monitor the system's operating status, promptly detect and handle faults, and evaluate the system's overall performance, significantly improving the monitoring, management, and operating efficiency of the central control system and ensuring the system's stability and security. The above technologies are all existing mature means and will not be elaborated on here.

[0183] The system evaluation module 34 provides an overall score for the central control system based on static data such as the current system status and fault information, as well as real-time status data, and proposes feasible suggestions for optimization analysis based on the current health status of each system module. The above technologies are all existing mature means and will not be elaborated on here.

[0184] The foregoing is merely a preferred embodiment of the present invention. It should be noted that those skilled in the art may make various improvements and modifications without departing from the principles of the present invention, and such improvements and modifications are also within the scope of protection of the present invention. Structures, devices, and operating methods not specifically described or explained herein shall, unless otherwise specified or limited, be implemented in accordance with conventional means in the art.

Claims

1. An intelligent monitoring and full-process simulation optimization system for an offshore oil platform central control system, comprising an offshore oil platform central control system (1), characterized in that: The offshore oil platform central control system (1) is internally provided with an intelligent monitoring platform (2) and a full-process simulation optimization platform (3); The internal configuration of the intelligent monitoring platform (2) includes: The abnormal history and analysis module (22) comprehensively analyzes the abnormal causes through multi-dimensional statistics, and is internally provided with an abnormal cause classification module (221) and an abnormal result analysis module (222); Fault prediction and optimization module (24) performs real-time simulation and fault prediction on the platform equipment operating status and process flow; The unmanned inspection integrated module (26) cooperates with the drones and cabins distributed in the corresponding areas to automatically inspect high-risk areas. The unmanned inspection integrated module (26) is internally provided with a drone control intervention module (261), a navigation interface module (262), an inspection video recording module (263), and a hovering inspection function module (264); The intelligent monitoring platform (2) is further provided with an equipment configuration monitoring module (21), an early warning statistics and suggestion module (23), and an abnormal alarm module (25); The equipment configuration monitoring module (21) is used to collect status signals of intelligent instruments and alarm signals of the platform environment; the equipment configuration monitoring module (21) is internally provided with an equipment status acquisition module (211) and a multi-dimensional data integration module (212); the equipment status acquisition module (211) classifies the status signals of the intelligent instruments into five levels according to the severity: no communication, fault, abnormality, maintenance, and suggestion; the multi-dimensional data integration module (212) integrates various abnormal environment alarm signals of the oil platform; The abnormality inducement classification module (221) is used to integrate and analyze abnormal device status signals and classify factors that cause device abnormalities; the abnormality result analysis module (222) provides corresponding countermeasures according to the type of abnormality inducement; Early warning statistics and suggestion module (23) is used to display the fault diagnosis information and historical events of the central control system alarm equipment in real time and provide guidance and suggestions; An abnormality alarm module (25) is used to generate an alarm when abnormal data is detected; The internal configuration of the full-process simulation optimization platform (3) includes: The full-process simulation optimization module (31) integrates the full-process semi-physical simulation technology and heterogeneous information processing methods, supports digital twin modeling, and performs real-time simulation of the platform equipment operating status and process flow; Smart instrument ledger module (32) systematically manages equipment ledgers and classifies and counts smart instruments based on equipment region, manufacturer, system, and equipment level; It also includes drones and cabins deployed in high-risk areas corresponding to offshore oil platforms, and both drones and cabins adopt an anti-corrosion fuselage design.

2. The intelligent monitoring and full-process simulation optimization system for the offshore oil platform central control system according to claim 1 is characterized by: The UAV control intervention module (261) performs human-machine collaborative inspection and data analysis through the generated control interface; The navigation interface module (262) three-dimensional path planning and generation of the UAV navigation route; The inspection video module (263) is used to track and record the inspection process in real time; The hovering inspection function module (264) develops a hovering inspection function to ensure that the UAV is in a relatively static state for inspection when the oil platform is shaking.

3. The intelligent monitoring and full-process simulation optimization system for the offshore oil platform central control system according to claim 1 is characterized by: The full-process simulation optimization platform (3) is further provided with a system architecture overview module (33) and a system evaluation module (34).

4. The intelligent monitoring and full-process simulation optimization system for the offshore oil platform central control system according to claim 3 is characterized by: The system architecture overview module (33) systematically introduces the system scale of the central control system, displays the overall architecture of the central control system, and dynamically displays the real-time online status of the operation station, the communication status of the central control system controller, and the communication status between the central control and third-party equipment.

5. The intelligent monitoring and full-process simulation optimization system for the offshore oil platform central control system according to claim 3 is characterized by: The system evaluation module (34) evaluates the overall score of the central control system based on the current system status, static data of fault information and real-time status data, and proposes feasible optimization suggestions.

6. The intelligent monitoring and full-process simulation optimization system for the offshore oil platform central control system according to claim 1 is characterized by: The full-process simulation optimization module (31) is internally provided with a semi-physical simulation module (311), a heterogeneous information processing module (312) and a digital twin modeling module (313); A semi-physical simulation module (311) is used to realize a full-process semi-physical simulation technology; a heterogeneous information processing module (312) is used to process and fuse heterogeneous information; Digital twin modeling and synchronization are performed through the digital twin modeling module (313).

Images