Offshore oil operation monitoring system and method based on visual identification

Abstract

The invention provides an offshore oil operation monitoring system and method based on visual identification, and belongs to the field of offshore oil operation monitoring. The method includes the steps that the first number of sensors of the same type with data collected by sensors exceeding a first threshold value and the second number of sensors of the same type with data exceeding a second threshold value are counted, and the first threshold value is smaller than the second threshold value; when the first number is larger than a first preset number or the second number is larger than a second preset number, it is preliminarily judged that an abnormal condition exists at the corresponding position, and the first preset number is larger than the second preset number; performing exception rechecking and positioning based on theoretical and actual data of the equipment and / or the parts; production parameters are automatically generated and adjusted in combination with data information, abnormal conditions and abnormal corresponding positions collected by different types of sensors; transferring automatic inspection equipment to carry out emergency inspection and operation and maintenance on the abnormal position; according to the invention, the accuracy of anomaly detection can be improved, and the operation and maintenance efficiency and safety are improved.

Description

Technical Field

[0001] The present invention relates to the field of offshore oil operation monitoring, and in particular to an offshore oil operation monitoring system and method based on visual recognition. Background Art

[0002] Offshore oil operations are related to national economic security and energy security. Due to the complex and changeable operating environment, offshore oil operations face great challenges and risks. In daily production, once there is a blowout, fire and explosion, platform distress, oil (gas) production facilities and pipeline damage / leakage, toxic and hazardous substance leakage, radioactive material escape; serious injury, death, disappearance, poisoning; oil spill accidents, natural disasters and other special circumstances, it will not only bring huge property losses, but also seriously threaten the personal safety of offshore workers. In order to effectively ensure the safety and stability of offshore oil operations, the offshore oil operation monitoring system has gradually become one of the essential tools for offshore oil operations. The offshore oil operation monitoring system realizes remote real-time monitoring of the production process and equipment facilities through sensors installed at various key nodes of the operation site. This monitoring system can help enterprise managers understand the production situation in real time, discover and solve problems in a timely manner, and improve production efficiency and safety.

[0003] The offshore oil operation monitoring system can realize real-time monitoring of the production process and equipment facilities based on the data information collected by the sensors, helping enterprise managers or production personnel to detect abnormal situations in time. The current offshore oil operation monitoring system faces at least one of the following problems: First, the existing technology often compares the data collected by a single sensor or the average value of the data collected by multiple sensors with the preset threshold to realize abnormal diagnosis. Its abnormal judgment method is relatively simple. This method is difficult to take into account the accurate monitoring of severe abnormalities in a small range and moderate abnormalities in a large range, and there are omissions or false alarms; for example, in a smaller working range, when there is a single sensor point data that is too high, if the abnormal diagnosis is realized based on the average value of the data collected by multiple sensors, there is a possibility of omission; in a larger working range, when there are multiple point sensor data that are high at the same time but have not reached a certain level, if the abnormal diagnosis is realized based on the data collected by a single sensor, there is also a possibility of omission; secondly, the sensor will produce accuracy drift and acquisition errors as the working time increases, especially In a complex operating environment, this precision drift and acquisition error will be more obvious, but accurate data collection based on sensors is the basis for accurate monitoring of the production process and equipment and facility status. How to accurately verify the reliability of sensors and achieve efficient operation and maintenance is a powerful measure to reduce operation and maintenance costs and improve operation and maintenance efficiency. In addition, increasing the density of sensor deployment is an effective measure to achieve accurate positioning of abnormalities, but it will bring about problems of excessive cost or installation difficulty. How to coordinate the deployment cost of sensors, installation difficulty and accurate positioning of abnormalities is a technical problem that needs to be solved urgently. Finally, in the process of inspecting the production process and equipment and facilities using unmanned automatic inspection equipment or manual inspection, in the face of a complex working environment, how to ensure the safety of personnel, unmanned automatic inspection equipment and production equipment and facilities is of great significance to ensuring safe production and reducing production costs. Summary of the invention

[0004] In view of this, the present invention aims to propose an offshore oil operation monitoring system and method based on visual recognition to solve the technical problems mentioned in the above background technology.

[0005] To achieve the above object, the technical solution of the present invention is implemented as follows: a method for monitoring offshore oil operations based on visual recognition, comprising the following steps:

[0006] S1: Use different types of sensors to collect input and output data of each device and / or component, as well as data related to intermediate facilities and the environment;

[0007] S2: Counting a first number of sensors of the same type whose collected data exceeds a first threshold, and a second number of sensors of the same type whose collected data exceeds a second threshold, wherein the first threshold is less than the second threshold;

[0008] S3: When the first number is greater than the first preset number or the second number is greater than the second preset number, it is preliminarily determined that an abnormality exists at the corresponding position, and the first preset number is greater than the second preset number;

[0009] S4: Review and locate the abnormality based on the input, theoretical output data and actual output data of the equipment and / or components, and determine the corresponding position of the abnormality;

[0010] S5: Analyze and process the data information collected by different types of sensors, and automatically generate and adjust the production parameters of equipment and / or components in the production process based on the abnormal conditions and the corresponding positions of the abnormalities;

[0011] S6: mobilize automatic inspection equipment to conduct emergency inspection and maintenance of abnormal locations;

[0012] S7: When the abnormal situation is serious, a manual operation and maintenance work order is generated, and the automatic inspection equipment and operation and maintenance personnel work together to carry out abnormal situation detection and operation and maintenance operations.

[0013] Furthermore, step S2 is specifically as follows: at the same time or in the same time period, counting the number of data collected by the same type of sensors exceeding the first threshold as the first quantity; and counting the number of data collected by the same type of sensors exceeding the second threshold as the second quantity.

[0014] Furthermore, in the step S3, the second preset number is 3, and a preliminary judgment is made that an abnormality exists at the corresponding position, including: mutual verification of abnormal conditions between different sensors, and when the data fluctuations of at least two sensors are consistent, a preliminary judgment is made that an abnormality exists at the corresponding position; otherwise, it is determined to be a false alarm, and a suspected fault mark is added to the sensor whose fluctuations are inconsistent with those of most sensors, and the automatic inspection equipment is mobilized to confirm it later.

[0015] Further, the step S4 includes:

[0016] S41: collecting input and output parameters of each device and / or component using sensors arranged at input and output positions of each device and / or component;

[0017] S42: Determine the input parameters of each device, and input them into the transfer function model corresponding to each device to obtain the theoretical output value of the device;

[0018] S43: taking the collected output parameters of each device as the actual output value of the device, and comparing the theoretical output value and the actual output value of each device;

[0019] S44: when the difference between the theoretical output value and the actual output value of the device is greater than a preset value, a review result is obtained, the device is determined to be abnormal, and the abnormal device location information is output; otherwise, it is determined to be a false alarm;

[0020] S45: determining whether the abnormal device includes a component, and if so, further determining the location information of the abnormal component;

[0021] S46: Determine the input parameters of each component, and input them into the transfer function model corresponding to each component to obtain the theoretical output value of the component;

[0022] S47: taking the collected output parameters of each component as the actual output value of the component, and comparing the theoretical output value and the actual output value of each component;

[0023] S48: When the difference between the theoretical output value and the actual output value of the component is greater than a preset value, a review result is obtained, the component is determined to be abnormal, and the abnormal component position information is output.

[0024] Furthermore, the step S5 comprises:

[0025] S51: preprocessing and feature extraction of the collected input and output data of each device and / or component, data of intermediate facilities, and data related to the surrounding environment to obtain a feature vector;

[0026] S52: Acquire the abnormal situation and the position corresponding to the abnormal situation, input the abnormal situation, the position corresponding to the abnormal situation and the characteristic vector into the preset control model, and the preset control module automatically generates and adjusts the production parameters of each device and / or component accordingly.

[0027] Another aspect of the present invention further provides an offshore oil operation monitoring system based on visual recognition, comprising: a control center, automatic inspection equipment, operation and maintenance auxiliary equipment, a preset container, equipment 1-n, different types of sensors, a visual recognition module, a cloud server, a mobile terminal, a first interactive module, a first display, and a first storage module; the control center is wirelessly connected to the automatic inspection equipment, the operation and maintenance auxiliary equipment, the preset container, the equipment 1-n, the temperature sensor, the pressure sensor, the flow sensor, the oil and gas concentration sensor, the wind speed sensor, the vibration frequency sensor, the stress sensor, the visual recognition module, the cloud server, the mobile terminal, the first interactive module, the first display, and the first storage module respectively;

[0028] The control center can use different types of sensors to collect input and output data of each device and / or component, data of intermediate facilities, and data related to the surrounding environment; count the first number of sensors of the same type whose collected data exceeds a first threshold, and the second number of sensors of the same type whose collected data exceeds a second threshold, and the first threshold is less than the second threshold; when the first number is greater than a first preset number or the second number is greater than a second preset number, preliminarily judge that there is an abnormality at the corresponding position, and the first preset number is greater than the second preset number; review and locate the abnormality based on the input, theoretical output data and actual output data of the device and / or component, and determine the corresponding position of the abnormality; analyze and process the data information collected by different types of sensors, and automatically generate and adjust the production parameters of the equipment and / or components in the production process in combination with the abnormal situation and the corresponding position of the abnormality; mobilize automatic inspection equipment to perform emergency inspection and operation and maintenance on the abnormal position; when the abnormal situation is serious, generate a manual operation and maintenance work order, and the automatic inspection equipment and operation and maintenance personnel jointly carry out abnormal situation detection and operation and maintenance operations;

[0029] Different types of sensors include temperature sensors, pressure sensors, flow sensors, oil and gas concentration sensors, wind speed sensors, vibration frequency sensors, and stress sensors. Different types of sensors are arranged at the input and output positions of each device or component according to their different functions to collect the input and output parameters of each device or component, or are arranged at the location of corresponding facilities to collect corresponding parameter information.

[0030] Furthermore, the control center can analyze and process the data information collected by different types of sensors, and automatically generate and adjust the production parameters of the equipment and / or components in the production process based on the abnormal situation and the corresponding position of the abnormality, including:

[0031] The collected input and output data of each device and / or component, the data of intermediate facilities, and the data related to the surrounding environment are preprocessed and feature extracted to obtain a feature vector; the abnormal situation and the corresponding position of the abnormality are obtained, and the abnormal situation, the corresponding position of the abnormality and the feature vector are input into the preset control model, and the preset control module automatically generates and adjusts the production parameters of each device and / or component accordingly.

[0032] Furthermore, the control center can analyze and process the data information collected by different types of sensors, automatically generate and adjust the production parameters of the equipment and / or components in the production process based on the abnormal situation and the corresponding position of the abnormality, and also include:

[0033] Obtain wind speed and earthquake warning data. When the wind speed exceeds the preset value or there is an earthquake warning higher than the preset level, stop production operations, call the automatic inspection equipment to return to the docking cabin, and start the lock to close the oil outlet;

[0034] The deformation size, vibration frequency and stress data at key positions of the equipment and / or components are obtained, the severity of the abnormality is determined in combination with the wind speed, and seawater is added to the preset container based on the severity of the abnormality.

[0035] Furthermore, the severity of the abnormality is determined in combination with the wind speed, and seawater is added to a preset container based on the severity of the abnormality, including:

[0036] Calculate the severity of the abnormality M = f(w, x, y, z), where f represents a preset function, w represents the deformation size, x represents the vibration frequency, y represents the stress data, and z represents the wind speed;

[0037] The height of the water level in the preset container is determined based on the severity of the abnormality, H=k×M, where k is a preset proportionality coefficient; and the water level height is added to H.

[0038] Furthermore, the system also includes a wave charging device, which is installed at an offshore oil operation site and is used to charge the automatic inspection equipment.

[0039] Compared with the prior art, the offshore oil operation monitoring system and method based on visual recognition described in the present invention has the following advantages:

[0040] The present invention provides an offshore oil operation monitoring system and method based on visual recognition, comprising using different types of sensors to collect input and output data of various equipment and / or components, data of intermediate facilities, and data related to the surrounding environment; counting a first number of sensors of the same type whose collected data exceeds a first threshold, and a second number of sensors of the same type whose collected data exceeds a second threshold, wherein the first threshold is less than the second threshold; when the first number is greater than a first preset number or the second number is greater than a second preset number, preliminarily judging that an abnormality exists at a corresponding position, wherein the first preset number is greater than the second preset number; reviewing and locating the abnormality based on the input, theoretical output data, and actual output data of the equipment and / or components, and determining the corresponding position of the abnormality; analyzing and processing the data information collected by different types of sensors, and automatically generating and adjusting the production parameters of the equipment and / or components in the production process in combination with the abnormality and the corresponding position of the abnormality; mobilizing automatic inspection equipment to perform emergency inspection and operation and maintenance on the abnormal position; when the abnormality is serious, generating a manual operation and maintenance work order, and the automatic inspection equipment and the operation and maintenance personnel jointly carry out the detection and operation and maintenance of the abnormality; the present invention has the following technical effects:

[0041] (1) By monitoring the number of sensors of the same type exceeding the preset threshold, it is possible to accurately monitor severe anomalies in a small area and moderate anomalies in a large area;

[0042] (2) Through mutual verification of abnormal conditions between different sensors, the reliability of sensors can be detected;

[0043] (3) Determine the time to overhaul equipment or components by calculating the difference between the theoretical output value and the actual output value of the equipment or components;

[0044] (4) Comprehensively consider equipment information, environmental information, and abnormal status information, and reasonably control the production of single wells according to the actual situation on site, and take measures to increase and stabilize production in a timely manner to maximize the production safety and oil and gas recovery efficiency of the oil and gas field and avoid property losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings:

[0046] Figure 1 It is a flowchart of the offshore oil operation monitoring method based on visual recognition;

[0047] Figure 2 This is a schematic diagram of the component structure of an underwater Christmas tree;

[0048] Figure 3 This is a structural diagram of the offshore oil operation monitoring system based on visual recognition. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. The specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in each embodiment of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0050] like Figure 1 As shown, the present invention is a method for monitoring offshore oil operations based on visual recognition, comprising the following steps:

[0051] S1: using different types of sensors to collect input and output data of each device and / or component, data of intermediate facilities, and data related to the surrounding environment; data related to the surrounding environment refers to data collected by sensors related to the environment, such as external wind speed, temperature, etc., which can also be called data related to the environment;

[0052] S2: Counting a first number of sensors of the same type whose collected data exceeds a first threshold, and a second number of sensors of the same type whose collected data exceeds a second threshold, wherein the first threshold is less than the second threshold;

[0053] S3: When the first number is greater than the first preset number or the second number is greater than the second preset number, it is preliminarily determined that an abnormality exists at the corresponding position, and the first preset number is greater than the second preset number;

[0054] The first preset number refers to the preset threshold of the number of sensors whose real-time collected data meets the condition of exceeding the first threshold, and the specific type is the number of sensors; the second preset number refers to the preset threshold of the number of sensors whose real-time collected data meets the condition of exceeding the second threshold, and the specific type is the number of sensors;

[0055] S4: Review and locate the abnormality based on the input, theoretical output data and actual output data of the equipment and / or components, and determine the corresponding position of the abnormality;

[0056] S5: Analyze and process the data information collected by different types of sensors, and automatically generate and adjust the production parameters of equipment and / or components in the production process based on the abnormal conditions and the corresponding positions of the abnormalities;

[0057] S6: mobilize automatic inspection equipment to conduct emergency inspection and maintenance of abnormal locations;

[0058] S7: When the abnormal situation is serious, a manual operation and maintenance work order is generated, and the automatic inspection equipment and operation and maintenance personnel work together to carry out abnormal situation detection and operation and maintenance operations.

[0059] In the description of the present invention, it is to be understood that the terms "first", "second", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", etc. may explicitly or implicitly include one or more of the feature. In the description of the present invention, unless otherwise specified, "plurality" means two or more.

[0060] In step S1, the offshore oil operation system includes multiple integrated modules of different functional links, and the device is an integrated module of a functional link in the offshore oil operation system. Therefore, multiple integrated modules correspond to multiple devices, and integrated modules with different functions correspond to different devices, such as underwater Christmas trees, underwater manifolds, water injection risers, oil risers, wellhead platforms, etc., and each device is connected in sequence through pipelines or corresponding fixed structures; each device is composed of different components, such as Figure 2 As shown, the underwater Christmas tree includes a Christmas tree cap, a top blocker, a tubing hanger, a valve group, an H4 connector, a wellhead and other components; the wellhead platform includes a base, a deck and columns and other components; this is only an example, and other equipment and corresponding components may also be included, which is not limited here.

[0061] The different types of sensors may be temperature sensors, pressure sensors, flow sensors, oil and gas concentration sensors, etc., which are arranged at the input and output positions of each device or component to collect the input and output parameters of each device or component; they may also be wind speed sensors, vibration frequency sensors, stress sensors, corrosion level sensors, which are arranged at the location of the corresponding facilities to collect the external wind speed, the vibration frequency of the equipment or platform, the stress magnitude, the corrosion degree, etc.; this is only an example, and other types of sensors may also be included according to the actual content to be monitored, which is not limited here;

[0062] When conducting offshore oil operations, abnormal situations are inevitable. The abnormal situations may be oil and gas leakage, pressure loss, etc. caused by cracks and gaps in the pipeline, or flow rate obstruction caused by corrosion. When an abnormal situation occurs, the parameter information collected by a few sensors in a small range may fluctuate greatly, or the parameter information collected by most sensors in a large range may fluctuate slightly. Traditional abnormality detection methods are difficult to accurately detect the two. To this end, the present invention proposes the following technical solutions:

[0063] Specifically, step S2 includes, at the same time or in the same time period, counting the number of data collected by the same type of sensors exceeding the first threshold as the first number; counting the number of data collected by the same type of sensors exceeding the second threshold as the second number; exceeding the first threshold corresponds to a smaller parameter information fluctuation; exceeding the second threshold corresponds to a larger parameter information fluctuation.

[0064] Correspondingly, in step S3, the first number is greater than the first preset number, corresponding to a relatively small range where the parameter information collected by a small number of sensors has undergone a large fluctuation; the second number is greater than the second preset number, corresponding to a relatively large range where the parameter information collected by a large number of sensors has undergone a small fluctuation; through the above method, the comprehensiveness and accuracy of anomaly detection can be improved;

[0065] The second preset number is greater than or equal to 1;

[0066] In one embodiment, the second preset number is 1; in this case, an abnormality review may be performed based on subsequent steps S4-S7;

[0067] In another embodiment, the second preset number is 2; at this time, the preliminary judgment that there is an abnormality at the corresponding position includes: mutual verification of the abnormal state between the two sensors, when the fluctuation of the data of the two sensors is consistent, it is preliminarily judged that there is an abnormality at the corresponding position; otherwise, it is judged as a false alarm, and suspected fault marks are added to the two sensors, and the automatic inspection equipment is mobilized to confirm it later;

[0068] In another embodiment, the second preset number is 3; at this time, the preliminary judgment that there is an abnormality at the corresponding position includes: mutual verification of abnormal states between different sensors, and when the fluctuations of at least two sensor data are consistent, it is preliminarily judged that there is an abnormality at the corresponding position; otherwise, it is judged as a false alarm, and a suspected fault mark is added to the sensor whose fluctuation is inconsistent with that of most sensors, and the automatic inspection equipment is mobilized to confirm it later;

[0069] Optionally, the second preset number may be greater than 3; in this case, the preliminary determination of the presence of an abnormality at the corresponding position is similar to the corresponding process when the second preset number is 3.

[0070] Through the above technical scheme, the present invention can avoid the omission and false detection caused by the single abnormality judgment method and sensor error to a certain extent, so that the accuracy of abnormality detection has been significantly improved, but how to further improve the accuracy of abnormality detection and how to quickly locate the abnormal position are still of great significance; generally speaking, increasing the density of sensor layout is an effective measure to achieve accurate positioning of abnormalities, but it will bring about the problem of excessive cost or installation difficulty. How to coordinate the accuracy of abnormality detection, the layout cost of sensors, the difficulty of installation and the accurate positioning of abnormalities are technical problems that need to be solved in this field; for this reason, the present invention proposes the following technical scheme:

[0071] The offshore oil operation system includes different equipment, each of which is connected in sequence; each equipment is composed of different components, each of which is also connected in sequence; different equipment and components correspond to different transfer function models, and the theoretical output value of the equipment or component can be determined based on the input parameters of the equipment or component and the transfer function model, for example, the theoretical output value of equipment m = input parameters of equipment m / (transfer function model of equipment m);

[0072] The step S4 comprises:

[0073] S41: collecting input and output parameters of each device and / or component using sensors arranged at input and output positions of each device and / or component;

[0074] S42: Determine the input parameters of each device, and input them into the transfer function model corresponding to each device to obtain the theoretical output value of the device;

[0075] S43: taking the collected output parameters of each device as the actual output value of the device, and comparing the theoretical output value and the actual output value of each device;

[0076] S44: when the difference between the theoretical output value and the actual output value of the device is greater than a preset value, a review result is obtained, the device is determined to be abnormal, and the abnormal device location information is output; otherwise, it is determined to be a false alarm;

[0077] S45: determining whether the abnormal device includes a component, and if so, further determining the location information of the abnormal component;

[0078] S46: Determine the input parameters of each component, and input them into the transfer function model corresponding to each component to obtain the theoretical output value of the component;

[0079] S47: taking the collected output parameters of each component as the actual output value of the component, and comparing the theoretical output value and the actual output value of each component;

[0080] S48: When the difference between the theoretical output value and the actual output value of the component is greater than a preset value, a review result is obtained, the component is determined to be abnormal, and the abnormal component position information is output.

[0081] Among them, in the step S42, the input parameters of each device can be: the input parameters of each device collected by the sensor in real time; for example, the offshore oil operation system includes three devices, namely device 1, device 2, and device 3, and the parameters collected in real time by the sensors respectively arranged at the input positions of device 1, device 2, and device 3 are used as the input parameters of device 1, device 2, and device 3 to calculate the theoretical output value of the device; the use of this method to calculate the theoretical output value of each device can avoid the accumulation of reasonable errors of each device itself, and realize accurate monitoring of the working status of each device;

[0082] Optionally, the input parameters of each device may also be: using the data collected by the sensor set at the input position of the first device as the input parameter of the first device, and using the theoretical output value of the device calculated by the first device input parameter and the transfer function model corresponding to each device as the input parameter of the next-level device; for example, the offshore oil operation system includes three devices, namely device 1, device 2, and device 3, and the three devices are connected in sequence, using the parameters collected in real time by the sensor set at the input position of device 1 as the input parameter of device 1, using the input parameters of device 1 and the transfer function model of device 1 to calculate the theoretical output value of device 1, and using the theoretical output value of device 1 as the input parameter of device 2, and subsequently using the theoretical output value of device 2 as the input parameter of device 3, to calculate the theoretical output value of the device; using this method to calculate the theoretical output value of each device can realize the monitoring of the overall state of the system; when using this method to calculate the theoretical output value of the device, it also includes:

[0083] When the difference between the theoretical output value and the actual output value of a certain level of equipment is within a first preset range, all equipment before the equipment of this level is subject to minor overhaul; when the difference between the theoretical output value and the actual output value of a certain level of equipment is within a second preset range, all equipment before the equipment of this level is subject to major overhaul; compared with the first preset range, the second preset range indicates that the overall error of the system is greater;

[0084] Similarly, in S46, the input parameters of each component may also be the two methods in S42; similar to the above scheme, a major overhaul or a minor overhaul of the system may also be performed based on the difference between the theoretical output value and the actual output value of a certain level of component;

[0085] The present invention can realize fast and accurate positioning of abnormalities while improving the accuracy of abnormality detection and reducing the cost and difficulty of sensor deployment through two-level review and positioning of equipment and components.

[0086] The step S5 comprises:

[0087] S51: preprocessing and feature extraction of the collected input and output data of each device and / or component, data of other intermediate facilities, and data related to the surrounding environment to obtain a feature vector;

[0088] S52: Acquire the abnormal situation and the position corresponding to the abnormal situation, input the abnormal situation, the position corresponding to the abnormal situation and the characteristic vector into the preset control model, and the preset control module automatically generates and adjusts the production parameters of each device and / or component accordingly.

[0089] The preset control model is obtained by training based on production data in the trial production stage; the production parameters of each device and / or component may be the opening and closing size of the valve, the suction size of the motor, etc.;

[0090] Optionally, also include:

[0091] S53: Obtain wind speed and earthquake warning data. When the wind speed exceeds a preset value or there is an earthquake warning higher than a preset level, stop production operations, call the automatic inspection equipment to return to the docking cabin, and start the locker to close the oil outlet;

[0092] S54: Obtain deformation magnitude, vibration frequency and stress data at key positions of equipment and / or components, determine the severity of the abnormality in combination with the wind speed, and inject seawater into a preset container based on the severity of the abnormality;

[0093] The deformation size is obtained based on the visual recognition module; the vibration frequency and stress data are obtained based on the corresponding sensors;

[0094] The method of determining the severity of the abnormality in combination with the wind speed and injecting seawater into a preset container based on the severity of the abnormality includes:

[0095] S541: Calculate the severity of the abnormality M=f(w, x, y, z), where f represents a preset function, w represents the deformation size, x represents the vibration frequency, y represents the stress data, and z represents the wind speed;

[0096] S542: Determine the height of the water level in the preset container based on the severity of the abnormality, H=k×M, where k is a preset proportionality coefficient;

[0097] S543: Add the water level to H;

[0098] The preset container refers to a buoyancy box fixed on an offshore oil operating platform; by filling the buoy with water to a preset height, the weight of the platform can be increased, thereby reducing the impact of severe weather such as strong winds, heavy rains and earthquakes.

[0099] The present invention can comprehensively consider equipment information, environmental information, and abnormal status information, and reasonably control the single well production according to the actual situation on site, and take measures to increase and stabilize production in a timely manner, so as to maximize the production safety and oil and gas recovery efficiency of the oil and gas field and avoid property losses.

[0100] S61: Determine the inspection route and frequency of the automatic inspection device in combination with the topological relationship and importance of each device and / or component, and the automatic inspection device performs inspection operations according to the inspection route and frequency;

[0101] S62: After completing the review and location of the abnormality, a path is planned based on the abnormal location, and the abnormal equipment or components at the abnormal location are inspected and maintained;

[0102] S63: When the sensor is found to be abnormal, determine whether it can complete the replacement task; if so, move itself to a suitable position and drive its own robotic arm to replace the sensor;

[0103] S64: If the sensor replacement task cannot be completed, the self-suspending device is activated, and the on-site data is continuously collected based on the self-sensor, and a query message is sent to the background to wait for further instructions, and data collection or inspection tasks are continued according to the reply instructions;

[0104] The replaceable sensors of the automatic inspection equipment refer to sensors that are detachably connected based on the magnetic suction module, which are easy to disassemble. The automatic inspection equipment itself carries a preset type of sensor and can realize the task of replacing the sensor based on the visual recognition module and the robotic arm;

[0105] The automatic inspection equipment includes an underwater robot and / or a water drone; the automatic inspection equipment itself is equipped with different types of sensors, and when performing automatic inspection, it can verify the accuracy and drift of sensors of the same type at fixed positions based on its own sensors; a hovering device is installed on the underwater robot body;

[0106] Optionally, a wave charging device is installed at the offshore oil operation site to facilitate charging of the automatic inspection equipment.

[0107] In step S7, the abnormal situation may be serious, such as a ruptured oil pipe, etc. Step S7 includes:

[0108] S71: The operation and maintenance personnel wear operation and maintenance auxiliary equipment and dive underwater, and call the underwater robot to guide the underwater path according to actual needs;

[0109] The operation and maintenance auxiliary device has a positioning module; the operation and maintenance auxiliary device also has a display module, which can view the abnormal status information of each device and / or component in real time; the operation and maintenance auxiliary device can also perform path planning and navigation, and display the planned path in real time in the display module; optionally, the operation and maintenance auxiliary device also has a hovering module, which the operation and maintenance personnel can turn on to perform hovering operations according to actual operation needs;

[0110] S72: A first traction module is provided on the underwater robot body, and the first traction module can be detachably connected to a second traction module of the operation and maintenance auxiliary equipment;

[0111] S73: With the help of the first traction module of the underwater robot, the operation and maintenance personnel can quickly reach the abnormal location to carry out operation and maintenance operations;

[0112] Optionally, also include:

[0113] S74: In an emergency, the operation and maintenance personnel can issue a rescue call command;

[0114] The emergency situation may be oxygen leakage, physical discomfort, etc., and emergency rescue is required at this time;

[0115] S75: The underwater robot responds to the rescue call command in real time, obtains the location information of the operation and maintenance personnel, and quickly moves to the corresponding position, and quickly brings the operation and maintenance personnel out of the water by connecting the first traction module and the second traction module;

[0116] Optionally, a water drone can be called. The water drone is provided with a third traction module. After the operation and maintenance personnel are quickly taken out of the water, the first traction module and the second traction module are separated, and the second traction module and the third traction module are connected; the operation and maintenance personnel can quickly reach the nearest safe position under the traction of the water drone;

[0117] The first traction module, the second traction module, and the third traction module are all detachably connected based on magnetic attraction;

[0118] Through the above technical solution, the present invention can achieve abnormally fast operation and maintenance, improve the safety of operation and maintenance personnel, and avoid the occurrence of personal accidents.

[0119] Another aspect of the present invention also provides an offshore oil operation monitoring system based on visual recognition, such as Figure 3 As shown, the system includes: a control center, automatic inspection equipment, operation and maintenance auxiliary equipment, a preset container, equipment 1-n, different types of sensors, a visual recognition module, a cloud server, a mobile terminal, a first interaction module, a first display, and a first storage module; the control center is wirelessly connected to the automatic inspection equipment, the operation and maintenance auxiliary equipment, the preset container, the equipment 1-n, the temperature sensor, the pressure sensor, the flow sensor, the oil and gas concentration sensor, the wind speed sensor, the vibration frequency sensor, the stress sensor, the visual recognition module, the cloud server, the mobile terminal, the first interaction module, the first display, and the first storage module respectively;

[0120] The control center can use different types of sensors to collect input and output data of each device and / or component, data of intermediate facilities, and data related to the surrounding environment; count the first number of sensors of the same type whose collected data exceeds a first threshold, and the second number of sensors of the same type whose collected data exceeds a second threshold, and the first threshold is less than the second threshold; when the first number is greater than a first preset number or the second number is greater than a second preset number, preliminarily judge that there is an abnormality at the corresponding position, and the first preset number is greater than the second preset number; review and locate the abnormality based on the input, theoretical output data and actual output data of the device and / or component, and determine the corresponding position of the abnormality; analyze and process the data information collected by different types of sensors, and automatically generate and adjust the production parameters of the equipment and / or components in the production process in combination with the abnormal situation and the corresponding position of the abnormality; mobilize automatic inspection equipment to perform emergency inspection and operation and maintenance on the abnormal position; when the abnormal situation is serious, generate a manual operation and maintenance work order, and the automatic inspection equipment and operation and maintenance personnel jointly carry out abnormal situation detection and operation and maintenance operations;

[0121] The first interaction module, the first display, and the first storage module are used to implement the functions of command interaction, data information display, and data information storage, respectively; regarding the connection and coordination relationship between them and other functional modules, those skilled in the art can set them according to actual needs, and no specific limitation is made here;

[0122] The automatic inspection equipment includes an underwater robot and / or an above-water drone;

[0123] Different types of sensors include temperature sensors, pressure sensors, flow sensors, oil and gas concentration sensors, wind speed sensors, vibration frequency sensors, and stress sensors. Different types of sensors are arranged at the input and output positions of each device or component according to their different functions to collect the input and output parameters of each device or component, or arranged at the location of the corresponding facility to collect the corresponding parameter information. Specifically, temperature sensors, pressure sensors, flow sensors, oil and gas concentration sensors, etc. are arranged at the input and output positions of each device or component to collect the input and output parameters of each device or component. Wind speed sensors, vibration frequency sensors, stress sensors, etc. are arranged at the location of the corresponding facility to collect external wind speed, vibration frequency of the equipment or platform, stress size, etc. Different types of sensors can also include corrosion level sensors, which are arranged at the location of the corresponding facility to collect the degree of corrosion of the equipment or platform; this is only an example, and other types of sensors may also be included according to the actual content to be monitored. The type, installation method, and installation location of the sensors can be selected and set according to actual needs, and are not limited here.

[0124] The temperature sensor can collect the temperature of the preset position of the equipment and / or components and the temperature of the external environment; the pressure sensor, flow sensor, oil and gas concentration sensor, wind speed sensor, vibration frequency sensor, stress sensor, and visual recognition module can collect data of the equipment and / or components, as well as other intermediate facilities and data related to the surrounding environment; the cloud server can provide earthquake early warning data; and the monitoring system can be remotely controlled and viewed through a mobile terminal;

[0125] The system also includes a wave charging device, which is installed at an offshore oil operation site to facilitate charging of the automatic inspection equipment;

[0126] The equipment 1-n is an integrated module of a functional link in the offshore oil operation system, such as an underwater Christmas tree, an underwater manifold, a water injection riser, an oil delivery riser, a wellhead platform, etc. The offshore oil operation system includes different equipment, each of which is connected in sequence through pipelines or corresponding fixed structures; each equipment is composed of different components, and each component is also connected in sequence;

[0127] The operation and maintenance auxiliary equipment refers to a wearable device that can assist the operation and maintenance personnel in performing operation and maintenance operations;

[0128] The preset container refers to a buoyancy box fixed on an offshore oil platform; by filling the buoy with a preset water level, the weight of the platform can be increased, thereby reducing the impact of severe weather such as strong winds, heavy rains and earthquakes;

[0129] The first number of sensors of the same type whose data collected exceeds a first threshold, and the second number of sensors of the same type whose data exceeds a second threshold, wherein the first threshold is less than the second threshold, refers to:

[0130] At the same time or in the same time period, the number of data collected by the same type of sensor exceeding the first threshold is counted as the first number; the number of data collected by the same type of sensor exceeding the second threshold is counted as the second number; exceeding the first threshold corresponds to a smaller parameter information fluctuation; exceeding the second threshold corresponds to a larger parameter information fluctuation;

[0131] Correspondingly, when the first number is greater than the first preset number or the second number is greater than the second preset number, it is preliminarily determined that an abnormality exists at the corresponding position, and the first preset number is greater than the second preset number, wherein the first number is greater than the first preset number, corresponding to a relatively small range, the parameter information collected by a small number of sensors has undergone a large fluctuation; the second number is greater than the second preset number, corresponding to a relatively large range, the parameter information collected by a large number of sensors has undergone a small fluctuation; through the above method, the comprehensiveness and accuracy of abnormality detection can be improved;

[0132] The second preset number is greater than or equal to 1;

[0133] In one embodiment, the second preset number is 1; in this case, an abnormality review may be performed based on subsequent steps S4-S7;

[0134] In another embodiment, the second preset number is 2; at this time, the preliminary judgment that there is an abnormality at the corresponding position includes: mutual verification of the abnormal state between the two sensors, when the fluctuation of the data of the two sensors is consistent, it is preliminarily judged that there is an abnormality at the corresponding position; otherwise, it is judged as a false alarm, and suspected fault marks are added to the two sensors, and the automatic inspection equipment is mobilized to confirm it later;

[0135] In another embodiment, the second preset number is 3; at this time, the preliminary judgment that there is an abnormality at the corresponding position includes: mutual verification of abnormal states between different sensors, and when the fluctuations of at least two sensor data are consistent, it is preliminarily judged that there is an abnormality at the corresponding position; otherwise, it is judged as a false alarm, and a suspected fault mark is added to the sensor whose fluctuation is inconsistent with that of most sensors, and the automatic inspection equipment is mobilized to confirm it later;

[0136] Optionally, the second preset number may be greater than 3; in this case, the preliminary determination of the presence of an abnormality at the corresponding position is similar to the corresponding process when the second preset number is 3;

[0137] The devices 1-n are connected in sequence, each device is composed of different components, and the components are also connected in sequence; different devices and components correspond to different transfer function models, and the theoretical output value of the device or component can be determined based on the input parameters of the device or component and the transfer function model;

[0138] The control center can review and locate the abnormality based on the input, theoretical output data and actual output data of the equipment and / or components, and determine the corresponding position of the abnormality, including:

[0139] Collecting input and output parameters of each device and / or component using sensors arranged at input and output locations of each device and / or component;

[0140] Determine the input parameters of each device and input them into the transfer function model corresponding to each device to obtain the theoretical output value of the device;

[0141] The collected output parameters of each device are used as the actual output value of the device, and the theoretical output value and the actual output value of each device are compared;

[0142] When the difference between the theoretical output value and the actual output value of the device is greater than the preset value, the review result is obtained, the device is determined to be abnormal, and the abnormal device location information is output, otherwise it is determined to be a false alarm;

[0143] Determine whether the abnormal device includes a component, and if so, further determine the location information of the abnormal component;

[0144] Determine the input parameters of each component and input them into the transfer function model corresponding to each component to obtain the theoretical output value of the component;

[0145] The collected output parameters of each component are used as the actual output value of the component, and the theoretical output value and the actual output value of each component are compared;

[0146] When the difference between the theoretical output value and the actual output value of a component is greater than a preset value, a review result is obtained, the component is determined to be abnormal, and the position information of the abnormal component is output;

[0147] The input parameters of each device are determined and input into the transfer function model corresponding to each device to obtain the theoretical output value of the device, wherein the input parameters of each device can be: the input parameters of each device collected by the sensor in real time;

[0148] Optionally, the input parameters of each device may also be: using data collected by a sensor arranged at an input position of a first device as a first device input parameter, and using the first device input parameter and a transfer function model corresponding to each device to calculate a theoretical output value of the device as an input parameter of a subsequent device; when using this method to calculate the theoretical output value of the device, it also includes:

[0149] When the difference between the theoretical output value and the actual output value of a certain level of equipment is within a first preset range, all equipment before the equipment of this level is subject to minor overhaul; when the difference between the theoretical output value and the actual output value of a certain level of equipment is within a second preset range, all equipment before the equipment of this level is subject to major overhaul; compared with the first preset range, the second preset range indicates that the overall error of the system is greater;

[0150] Similarly, the input parameters of each component are determined and input into the transfer function model corresponding to each component to obtain the theoretical output value of the component; wherein the input parameters of each component can also be the above two methods; similar to the above scheme, a major overhaul or minor overhaul of the system can also be performed based on the difference between the theoretical output value and the actual output value of a certain level of component;

[0151] The data information collected by different types of sensors is analyzed and processed, and the production parameters of the equipment and / or components in the production process are automatically generated and adjusted in combination with the abnormal situation and the corresponding position of the abnormality, including: pre-processing and feature extraction of the collected input and output data of each device and / or component and the data of other intermediate facilities and the data related to the surrounding environment to obtain a feature vector; obtaining the abnormal situation and the corresponding position of the abnormality, and inputting the abnormal situation, the corresponding position of the abnormality and the feature vector into a preset control model, and the preset control module automatically generates and adjusts the production parameters of each device and / or component accordingly;

[0152] The preset control model is obtained by training based on production data in the trial production stage; the production parameters of each device and / or component may be the opening and closing size of the valve, the suction size of the motor, etc.;

[0153] Optionally, the analyzing and processing of data information collected by different types of sensors, automatically generating and adjusting production parameters of equipment and / or components in the production process in combination with abnormal conditions and corresponding positions of the abnormalities, also includes:

[0154] Obtain wind speed and earthquake warning data. When the wind speed exceeds the preset value or there is an earthquake warning higher than the preset level, stop production operations, call the automatic inspection equipment to return to the docking cabin, and start the lock to close the oil outlet;

[0155] Obtain deformation size, vibration frequency and stress data at key locations of equipment and / or components, determine the severity of the abnormality in combination with the wind speed, and add seawater to a preset container based on the severity of the abnormality;

[0156] The method of determining the severity of the abnormality in combination with the wind speed and injecting seawater into a preset container based on the severity of the abnormality includes:

[0157] Calculate the severity of the abnormality M=f(w, x, y, z), where f represents a preset function, w represents the deformation size, x represents the vibration frequency, y represents the stress data, and z represents the wind speed; determine the height of the water level in the preset container based on the severity of the abnormality, H=k×M, where k is a preset proportional coefficient; add the water level height to H;

[0158] The control center can mobilize automatic inspection equipment to conduct emergency inspections and maintenance of abnormal locations, including:

[0159] In combination with the topological relationship and importance of each device and / or component, the inspection route and frequency of the automatic inspection device are determined, and the automatic inspection device performs inspection operations according to the inspection route and frequency;

[0160] After completing the review and location of the abnormality, the path is planned based on the abnormal location, and the abnormal equipment or components at the abnormal location are inspected and maintained;

[0161] When a sensor is found to be abnormal, it determines whether it can complete the replacement task; if so, it moves itself to a suitable position and drives its own robotic arm to replace the sensor;

[0162] If the sensor replacement task cannot be completed, the robot will start its own hovering device and continue to collect on-site data based on its own sensors, send a query message to the background to wait for further instructions, and continue to collect data or continue to perform inspection tasks according to the reply instructions;

[0163] The replaceable sensor of the automatic inspection device refers to a sensor that is detachably connected based on a magnetic suction module, which has the characteristic of being easy to disassemble. The automatic inspection device itself carries a preset type of sensor and can realize the task of replacing the sensor based on a visual recognition module and a robotic arm;

[0164] The automatic inspection equipment itself is equipped with different types of sensors. When performing automatic inspection, it can verify the accuracy and drift of sensors of the same type at fixed positions based on its own sensors. A hovering device is installed on the underwater robot body.

[0165] When the abnormal situation is serious, the control center can generate a manual operation and maintenance work order, and the automatic inspection equipment and operation and maintenance personnel will work together to detect the abnormal situation and carry out operation and maintenance operations; including:

[0166] The operation and maintenance personnel wear the operation and maintenance auxiliary equipment and dive underwater, and call the underwater robot to guide the underwater path according to actual needs; the underwater robot body is provided with a first traction module, and the first traction module can be detachably connected with the second traction module of the operation and maintenance auxiliary equipment; the operation and maintenance personnel can quickly reach the abnormal position to carry out the operation and maintenance operation with the help of the first traction module of the underwater robot;

[0167] The operation and maintenance auxiliary device has a positioning module; the operation and maintenance auxiliary device also has a display module, which can view the abnormal status information of each device and / or component in real time; the operation and maintenance auxiliary device can also perform path planning and navigation, and display the planned path in real time in the display module; optionally, the operation and maintenance auxiliary device also has a hovering module, which the operation and maintenance personnel can turn on to perform hovering operations according to actual operation needs;

[0168] Optionally, the control center can generate a manual operation and maintenance work order when the abnormal situation is serious, and the automatic inspection equipment and operation and maintenance personnel will work together to detect the abnormal situation and perform operation and maintenance operations, including:

[0169] In an emergency, the operation and maintenance personnel can issue a rescue call command; the underwater robot responds to the rescue call command in real time, obtains the location information of the operation and maintenance personnel, and quickly moves to the corresponding position, and quickly brings the operation and maintenance personnel out of the water by connecting the first traction module and the second traction module;

[0170] Optionally, a water drone can be called. The water drone is provided with a third traction module. After the operation and maintenance personnel are quickly taken out of the water, the first traction module and the second traction module are separated, and the second traction module and the third traction module are connected; the operation and maintenance personnel can quickly reach the nearest safe position under the traction of the water drone;

[0171] The first traction module, the second traction module, and the third traction module are all detachably connected based on magnetic attraction;

[0172] Through the above technical solution, the present invention can achieve abnormally fast operation and maintenance, improve the safety of operation and maintenance personnel, and avoid the occurrence of personal accidents.

[0173] The present invention provides an offshore oil operation monitoring system and method based on visual recognition, comprising: using different types of sensors to collect input and output data of various equipment and / or components, data of intermediate facilities, and data related to the surrounding environment; counting a first number of sensors of the same type whose collected data exceeds a first threshold, and a second number of sensors of the same type whose collected data exceeds a second threshold, wherein the first threshold is less than the second threshold; when the first number is greater than a first preset number or the second number is greater than a second preset number, preliminarily judging that an abnormality exists at a corresponding position, wherein the first preset number is greater than the second preset number; reviewing and locating the abnormality based on the input, theoretical output data, and actual output data of the equipment and / or components, and determining the corresponding position of the abnormality; analyzing and processing the data information collected by different types of sensors, and automatically generating and adjusting the production parameters of the equipment and / or components in the production process in combination with the abnormality and the corresponding position of the abnormality; mobilizing automatic inspection equipment to perform emergency inspection and operation and maintenance on the abnormal position; when the abnormality is serious, generating a manual operation and maintenance work order, and the automatic inspection equipment and the operation and maintenance personnel jointly carry out the detection and operation and maintenance of the abnormality; the present invention can improve the accuracy of abnormality detection, and increase the efficiency and safety of operation and maintenance.

[0174] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for monitoring offshore oil operations based on visual recognition, characterized in that: The following steps are involved: S1: Use different types of sensors to collect input and output data of each device and / or component, data of intermediate facilities, and data related to the surrounding environment; S2: Counting a first number of sensors of the same type whose collected data exceeds a first threshold, and a second number of sensors of the same type whose collected data exceeds a second threshold, wherein the first threshold is less than the second threshold; S3: When the first number is greater than the first preset number or the second number is greater than the second preset number, it is preliminarily determined that an abnormality exists at the corresponding position, and the first preset number is greater than the second preset number; S4: Review and locate the abnormality based on the input, theoretical output data and actual output data of the equipment and / or components, and determine the corresponding position of the abnormality; S5: Analyze and process the data information collected by different types of sensors, and automatically generate and adjust the production parameters of equipment and / or components in the production process based on the abnormal conditions and the corresponding positions of the abnormalities; S6: mobilize automatic inspection equipment to conduct emergency inspection and maintenance of abnormal locations; S7: When the abnormal situation is serious, a manual operation and maintenance work order is generated, and the automatic inspection equipment and operation and maintenance personnel work together to carry out abnormal situation detection and operation and maintenance operations.

2. The method for monitoring offshore oil operations based on visual recognition according to claim 1, characterized in that: Specifically, step S2 includes, at the same time or in the same time period, counting the number of data collected by the same type of sensors exceeding the first threshold as the first quantity; and counting the number of data collected by the same type of sensors exceeding the second threshold as the second quantity.

3. The method for monitoring offshore oil operations based on visual recognition according to claim 2, characterized in that: In the step S3, the second preset number is 3, and a preliminary judgment is made that an abnormality exists at the corresponding position, including: mutual verification of abnormal conditions between different sensors, and when the data fluctuations of at least two sensors are consistent, a preliminary judgment is made that an abnormality exists at the corresponding position; otherwise, it is determined to be a false alarm, and a suspected fault mark is added to the sensor whose fluctuation is inconsistent with that of most sensors, and the automatic inspection equipment is mobilized to confirm it later.

4. The method for monitoring offshore oil operations based on visual recognition according to claim 3, characterized in that: The step S4 comprises: S41: collecting input and output parameters of each device and / or component using sensors arranged at input and output positions of each device and / or component; S42: Determine the input parameters of each device, and input them into the transfer function model corresponding to each device to obtain the theoretical output value of the device; S43: taking the collected output parameters of each device as the actual output value of the device, and comparing the theoretical output value and the actual output value of each device; S44: when the difference between the theoretical output value and the actual output value of the device is greater than a preset value, a review result is obtained, the device is determined to be abnormal, and the abnormal device location information is output; otherwise, it is determined to be a false alarm; S45: determining whether the abnormal device includes a component, and if so, further determining the location information of the abnormal component; S46: Determine the input parameters of each component, and input them into the transfer function model corresponding to each component to obtain the theoretical output value of the component; S47: taking the collected output parameters of each component as the actual output value of the component, and comparing the theoretical output value and the actual output value of each component; S48: When the difference between the theoretical output value and the actual output value of the component is greater than a preset value, a review result is obtained, the component is determined to be abnormal, and the abnormal component position information is output.

5. The method for monitoring offshore oil operations based on visual recognition according to claim 4, characterized in that: The step S5 comprises: S51: preprocessing and feature extraction of the collected input and output data of each device and / or component, data of intermediate facilities, and data related to the surrounding environment to obtain a feature vector; S52: Acquire the abnormal situation and the position corresponding to the abnormal situation, input the abnormal situation, the position corresponding to the abnormal situation and the characteristic vector into the preset control model, and the preset control module automatically generates and adjusts the production parameters of each device and / or component accordingly.

6. An offshore oil operation monitoring system based on visual recognition based on the method according to any one of claims 1 to 5, characterized in that: include: Control center, automatic inspection equipment, operation and maintenance auxiliary equipment, preset container, equipment 1-n, different types of sensors, visual recognition module, cloud server, mobile terminal, first interaction module, first display, first storage module; the control center is wirelessly connected with automatic inspection equipment, operation and maintenance auxiliary equipment, preset container, equipment 1-n, temperature sensor, pressure sensor, flow sensor, oil and gas concentration sensor, wind speed sensor, vibration frequency sensor, stress sensor, visual recognition module, cloud server, mobile terminal, first interaction module, first display, and first storage module respectively; The control center can use different types of sensors to collect input and output data of each device and / or component, data of intermediate facilities, and data related to the surrounding environment; Counting a first number of sensors of the same type whose collected data exceeds a first threshold, and a second number of sensors of the same type whose collected data exceeds a second threshold, wherein the first threshold is less than the second threshold; when the first number is greater than a first preset number or the second number is greater than a second preset number, preliminarily determining that an abnormality exists at the corresponding position, and the first preset number is greater than the second preset number; Review and locate abnormalities based on the input, theoretical output data, and actual output data of the equipment and / or components, and determine the corresponding location of the abnormality; Analyze and process the data information collected by different types of sensors, automatically generate and adjust the production parameters of equipment and / or components in the production process based on abnormal conditions and corresponding abnormal locations; mobilize automatic inspection equipment to conduct emergency inspections and operation and maintenance of abnormal locations; when the abnormal conditions are serious, generate manual operation and maintenance work orders, and the automatic inspection equipment and operation and maintenance personnel will work together to detect abnormal conditions and carry out operation and maintenance operations; Different types of sensors include temperature sensors, pressure sensors, flow sensors, oil and gas concentration sensors, wind speed sensors, vibration frequency sensors, and stress sensors. Different types of sensors are arranged at the input and output positions of each device or component according to their different functions to collect the input and output parameters of each device or component, or are arranged at the location of corresponding facilities to collect corresponding parameter information.

7. The offshore oil operation monitoring system based on visual recognition according to claim 6 is characterized in that: The control center can analyze and process the data information collected by different types of sensors, and automatically generate and adjust the production parameters of equipment and / or components in the production process based on abnormal conditions and their corresponding locations, including: The collected input and output data of each device and / or component, the data of the intermediate facilities, and the data related to the surrounding environment are preprocessed and feature extracted to obtain a feature vector; the abnormal situation and the corresponding position of the abnormality are obtained, and the abnormal situation, the corresponding position of the abnormality and the feature vector are input into the preset control model, and the preset control module automatically generates and adjusts the production parameters of each device and / or component accordingly.

8. The offshore oil operation monitoring system based on visual recognition according to claim 6 is characterized in that: The control center can analyze and process the data information collected by different types of sensors, automatically generate and adjust the production parameters of equipment and / or components in the production process based on abnormal conditions and their corresponding locations, and also includes: Obtain wind speed and earthquake warning data. When the wind speed exceeds the preset value or there is an earthquake warning higher than the preset level, stop production operations, call the automatic inspection equipment to return to the docking cabin, and start the lock to close the oil outlet; The deformation size, vibration frequency and stress data at key positions of the equipment and / or components are obtained, the severity of the abnormality is determined in combination with the wind speed, and seawater is added to the preset container based on the severity of the abnormality.

9. The offshore oil operation monitoring system based on visual recognition according to claim 8 is characterized in that: The severity of the abnormality is determined in combination with the wind speed, and seawater is added to the preset container based on the severity of the abnormality, including: Calculate the severity of the abnormality M = f(w, x, y, z), where f represents a preset function, w represents the deformation size, x represents the vibration frequency, y represents the stress data, and z represents the wind speed; The height of the water level in the preset container is determined based on the severity of the abnormality, H=k×M, where k is a preset proportionality coefficient; and the water level height is added to H.

10. The offshore oil operation monitoring system based on visual recognition according to claim 6, characterized in that: The system also includes a wave charging device, which is installed at an offshore oil operation site and is used to charge the automatic inspection equipment.

Images