CN117408502B

CN117408502B - Data stream arrangement method and system applied to oil and gas production system

Info

Publication number: CN117408502B
Application number: CN202311727435.1A
Authority: CN
Inventors: 王荣德; 李洪浪; 吴凯佳; 曹洋; 周杨波; 岳杰; 何春; 朱君; 任晓翠; 任康
Original assignee: Chengdu Chuanyou Ruifei Technology Co ltd
Current assignee: Chengdu Chuanyou Ruifei Technology Co ltd
Priority date: 2023-12-15
Filing date: 2023-12-15
Publication date: 2024-03-15
Anticipated expiration: 2043-12-15
Also published as: CN117408502A

Abstract

The invention provides a data flow arrangement method and a system applied to an oil gas production system, and relates to the technical field of artificial intelligence. In the invention, the equipment operation state data is subjected to feature mining, and the equipment operation state features corresponding to the equipment operation state data are output; according to the equipment running state characteristics, subsystem information matching is carried out on the equipment running state data, and corresponding matching storage subsystem information of the equipment running state data in the target storage frame information is output; according to the equipment running state characteristics and the storage equipment characteristics of each piece of storage equipment information, equipment running state data are subjected to equipment information matching, and corresponding matched storage equipment information of the equipment running state data in the target storage frame information is output; and arranging the equipment running state data based on the corresponding matched storage equipment information. Based on the above, the problem of relatively low reliability of the arrangement of the data streams can be improved.

Description

Data stream arrangement method and system applied to oil and gas production system

Technical Field

The invention relates to the technical field of artificial intelligence, in particular to a data flow arrangement method and system applied to an oil gas production system.

Background

The data flow arrangement method of the oil and gas production system is to plan, organize and control the data and the process involved in the system so as to realize efficient and reliable data flow and processing. The following are several common data stream scheduling methods applied to oil and gas production systems:

event driven architecture: many events in oil and gas production systems require triggering corresponding data processing operations; the event-driven architecture realizes automatic circulation and processing of data in the system by defining events and corresponding processing programs; for example, when a certain sensor detects an abnormal condition, a corresponding event is triggered, and the system automatically performs operations such as data acquisition, analysis, alarm and the like;

data pipeline: a data pipe is a way to transfer data from one link to another; in an oil and gas production system, a data pipeline can be used for connecting different data sources, processing programs and output targets to form a complete data flow path; the data pipeline can be configured and adjusted according to the requirements, so that data can flow according to a preset sequence and mode;

and (3) distributed storage: because the oil gas production system involves the storage of a large amount of data, the distributed storage can improve the efficiency and the searching convenience of the system; by dividing the data stream into a plurality of storage tasks and executing the tasks in parallel on different storage nodes, the data processing speed can be increased.

In this case, the data stream is divided into a plurality of tasks, and a large amount of data in the oil and gas production system is required to be divided to form a plurality of sub-data streams, however, in the prior art, division arrangement is generally performed directly based on time or source, so that the arrangement reliability is not high.

Disclosure of Invention

In view of the above, the present invention is directed to a data stream arranging method and system for an oil and gas production system, so as to improve the problem of relatively low reliability of data stream arrangement.

In order to achieve the above purpose, the embodiment of the present invention adopts the following technical scheme:

a data stream orchestration method applied to an oil and gas production system, comprising:

acquiring equipment operation state data of oil gas production equipment in a target oil gas production system, carrying out feature mining on the equipment operation state data by utilizing a feature mining unit included in a monitoring data matching network, and outputting equipment operation state features corresponding to the equipment operation state data, wherein the equipment operation state data is used for describing an operation process of the oil gas production equipment;

utilizing a first data matching unit included in the monitoring data matching network to match subsystem information of the equipment operation state data according to the equipment operation state characteristics, outputting matched storage subsystem information corresponding to the equipment operation state data in target storage frame information, wherein the target storage frame information comprises at least two storage subsystem information and a plurality of storage equipment information formed by subdividing the at least two storage subsystem information, each storage equipment information comprises at least one representative storage data segment, the storage subsystem information is used for reflecting one storage subsystem in a storage system of the equipment operation state data, each storage subsystem comprises at least two storage equipment, each storage equipment corresponds to one storage equipment information, and at least one representative storage data segment included in the storage equipment information is used for representing historical equipment operation state data stored by the corresponding storage equipment;

Performing device information matching on the device operation state data according to the device operation state characteristics and storage device characteristics of each piece of storage device information by using a second data matching unit included in the monitoring data matching network, and outputting matched storage device information corresponding to the device operation state data in the target storage frame information, wherein the storage device characteristics are determined according to representative storage data fragments included in corresponding storage device information, and the matched storage device information belongs to one storage device information of at least two storage device information corresponding to the matched storage subsystem information;

and arranging the equipment operation state data based on the corresponding matched storage equipment information, so that the equipment operation state data is stored in the storage equipment corresponding to the matched storage equipment information.

In some preferred embodiments, in the above data stream arranging method applied to an oil and gas production system, the second data matching unit includes a feature aggregation subunit and a data matching subunit;

the step of using the second data matching unit included in the monitoring data matching network to match the equipment operation state data with the storage equipment characteristics of each storage equipment information according to the equipment operation state characteristics and the storage equipment characteristics of each storage equipment information, and outputting the matched storage equipment information corresponding to the equipment operation state data in the target storage frame information includes:

Utilizing the characteristic aggregation subunit to aggregate the running state characteristics of the equipment and the storage equipment characteristics corresponding to the storage equipment information to form corresponding aggregation characteristics;

and carrying out equipment information matching on the equipment operation state data according to the aggregation characteristics by utilizing the data matching subunit, and outputting corresponding matched storage equipment information of the equipment operation state data in the target storage frame information.

In some preferred embodiments, in the above data stream arrangement method applied to an oil and gas production system, the step of using the feature aggregation subunit to aggregate the device running state feature and the storage device feature corresponding to each piece of storage device information to form a corresponding aggregate feature includes:

determining a first modality linear mapping parameter for performing a first modality linear mapping;

performing first-mode linear mapping on the equipment running state characteristics and the storage equipment characteristics of each piece of storage equipment information according to the first-mode linear mapping parameters by using the characteristic aggregation subunit to form corresponding first-mode linear mapping characteristics;

Determining a second modality linear mapping parameter for performing a second modality linear mapping;

performing second-mode linear mapping on the equipment running state characteristics and the storage equipment characteristics of each piece of storage equipment information according to the second-mode linear mapping parameters by using the characteristic aggregation subunit to form corresponding second-mode linear mapping characteristics;

and aggregating the first modality linear mapping feature and the second modality linear mapping feature by using the feature aggregation subunit to form a corresponding aggregation feature.

determining a first modality linear mapping parameter for performing a first modality linear mapping; and performing first-modality linear mapping on the device running state characteristics and the storage device characteristics of each piece of storage device information according to the first-modality linear mapping parameters by using the characteristic aggregation subunit to form corresponding first-modality linear mapping characteristics; and obtaining corresponding aggregation features based on the first modality linear mapping features; or alternatively

Determining a second modality linear mapping parameter for performing a second modality linear mapping; and performing second-modality linear mapping on the device running state characteristics and the storage device characteristics of each piece of storage device information according to the second-modality linear mapping parameters by using the characteristic aggregation subunit to form corresponding second-modality linear mapping characteristics; and obtaining corresponding aggregation features based on the second modality linear mapping features.

In some preferred embodiments, in the above data stream arrangement method applied to an oil and gas production system, before the step of performing subsystem information matching on the equipment operation state data according to the equipment operation state characteristics by using the first data matching unit included in the monitoring data matching network, and outputting matching storage subsystem information corresponding to the equipment operation state data in the target storage frame information, the data stream arrangement method applied to an oil and gas production system further includes:

determining representative storage data fragments included in each piece of storage equipment information in the target storage frame information;

for each piece of storage equipment information, performing feature mining on each representative storage data segment included in the storage equipment information to form a representative storage data segment feature corresponding to each representative storage data segment;

And carrying out feature aggregation on the storage device information including the features of the representative storage data fragments corresponding to the representative storage data fragments to form storage device features corresponding to the storage device information.

In some preferred embodiments, in the above data stream arranging method applied to an oil and gas production system, the step of feature-aggregating the storage device information including the representative storage data segment features corresponding to the representative storage data segment to form the storage device features corresponding to the storage device information includes:

for each piece of storage equipment information, when the number of fragments of the representative storage data fragments included in the storage equipment information is greater than or equal to the predetermined reference fragment number, carrying out averaging processing on the characteristic of the representative storage data fragments corresponding to the representative storage data fragments included in the storage equipment information to form the storage equipment characteristic corresponding to the storage equipment information;

and for each piece of storage equipment information, performing feature superposition processing on the storage equipment information including the characteristic of the representative storage data fragment corresponding to the representative storage data fragment to form the storage equipment feature corresponding to the storage equipment information under the condition that the number of the representative storage data fragment included in the storage equipment information is smaller than the number of the reference fragments.

In some preferred embodiments, in the above data stream arrangement method applied to an oil gas production system, the step of obtaining the equipment operation state data of the oil gas production equipment in the target oil gas production system, and using a feature mining unit included in the monitoring data matching network to perform feature mining on the equipment operation state data, and outputting the equipment operation state feature corresponding to the equipment operation state data includes:

acquiring equipment operation state data of oil gas production equipment in a target oil gas production system;

determining representative data segments of the equipment operation state data to form at least one state representative data segment included in the equipment operation state data;

the characteristic mining unit is used for respectively performing characteristic mining on all state representing data fragments included in the equipment operation state data and outputting state representing data fragment characteristics corresponding to all state representing data fragments;

and aggregating the output states to represent the data segment characteristics and forming the equipment operation state characteristics corresponding to the equipment operation state data.

In some preferred embodiments, in the above data stream arrangement method applied to an oil and gas production system, the step of using the first data matching unit included in the monitoring data matching network to match subsystem information of the equipment operation state data according to the equipment operation state feature, and outputting matching storage subsystem information corresponding to the equipment operation state data in target storage frame information includes:

Determining the at least two storage subsystem information included in the target storage frame information;

using a first data matching unit included in the monitoring data matching network, according to the equipment operation state characteristics and the at least two pieces of storage subsystem information, performing data matching on the equipment operation state data by the at least two pieces of storage subsystem information, and outputting a possibility parameter that the equipment operation state data is matched with each piece of storage subsystem information;

and marking the storage subsystem information corresponding to the possibility parameter with the maximum value to form the matched storage subsystem information corresponding to the equipment running state data.

In some preferred embodiments, in the above data stream arrangement method applied to an oil and gas production system, before the step of obtaining the device operation state data of the oil and gas production device in the target oil and gas production system and using the feature mining unit included in the monitoring data matching network to perform feature mining on the device operation state data and output the device operation state feature corresponding to the device operation state data, the data stream arrangement method applied to an oil and gas production system further includes:

Extracting operation state data of training equipment, wherein the operation state data of the training equipment is configured with real storage subsystem information and real storage equipment information;

performing feature mining on the training equipment operation state data by using the feature mining unit, and outputting training equipment operation state features corresponding to the training equipment operation state data;

using the first data matching unit to match subsystem information of the training equipment operation state data according to the training equipment operation state characteristics, and outputting corresponding subsystem information matching data, wherein the subsystem information matching data is used for reflecting matched storage subsystem information estimated by the training equipment operation state data;

using the second data matching unit to perform equipment information matching on the training equipment operation state data according to the training equipment operation state characteristics and the storage equipment characteristics of each piece of storage equipment information, and outputting corresponding equipment information matching data, wherein the equipment information matching data is used for reflecting matched storage equipment information estimated by the training equipment operation state data;

determining subsystem information differences between the real storage subsystem information and the subsystem information matching data, and determining storage device information differences between the real storage device information and the device information matching data;

Based on the subsystem information distinction and the storage device information distinction, adjusting network parameters of the monitoring data matching network to form a trained monitoring data matching network;

the step of adjusting the network parameters of the monitoring data matching network based on the subsystem information distinction and the storage device information distinction to form a trained monitoring data matching network comprises the following steps:

determining a first error index calculation rule corresponding to the first data matching unit, a second error index calculation rule corresponding to the second data matching unit and a dependency error index calculation rule, wherein the dependency error index calculation rule is used for constraining storage equipment information reflected by the equipment information matching data from being subordinate to storage subsystem information reflected by the subsystem information matching data;

determining a target error index calculation rule of the monitoring data matching network according to the first error index calculation rule, the second error index calculation rule and the dependency error index calculation rule;

calculating a target error index of the monitoring data matching network based on the target error index calculation rule according to the subsystem information difference and the storage device information difference;

And adjusting network parameters of the monitoring data matching network according to the target error index to form a trained monitoring data matching network.

The embodiment of the invention also provides a data stream arranging system which comprises a processor and a memory, wherein the memory is used for storing a computer program, and the processor is used for executing the computer program so as to realize the data stream arranging method applied to the oil gas production system.

The data flow arrangement method and the data flow arrangement system applied to the oil and gas production system provided by the embodiment of the invention can perform characteristic mining on the equipment operation state data and output the equipment operation state characteristics corresponding to the equipment operation state data; according to the equipment running state characteristics, subsystem information matching is carried out on the equipment running state data, and corresponding matching storage subsystem information of the equipment running state data in the target storage frame information is output; according to the equipment running state characteristics and the storage equipment characteristics of each piece of storage equipment information, equipment running state data are subjected to equipment information matching, and corresponding matched storage equipment information of the equipment running state data in the target storage frame information is output; and arranging the equipment running state data based on the corresponding matched storage equipment information. Based on the foregoing, on the one hand, because two-level data matching is performed, the accuracy of data matching is higher, on the other hand, because the storage device characteristics of the storage device information corresponding to the device running state data are determined by representing the storage data segments, and the representing storage data segments can represent the semantics expressed by the storage device information to a certain extent, the storage device characteristics corresponding to the storage device information and the device running state characteristics of the device running state data are determined based on the representing storage data segments of the storage device information, the device running state data are subjected to device information matching, the obtained matching storage device information to which the device running state data belongs can be more accurate, and the accuracy of data matching for the device running state data can be improved, so that the problem that the reliability of data stream arrangement in the prior art is relatively not high is solved.

In order to make the above objects, features and advantages of the present invention more comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

Fig. 1 is a block diagram of a data stream arranging system according to an embodiment of the present invention.

FIG. 2 is a flow chart of steps involved in a data flow arrangement method for an oil and gas production system according to an embodiment of the present invention.

FIG. 3 is a schematic diagram of each module included in a data flow arrangement apparatus for an oil and gas production system according to an embodiment of the present invention.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.

Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

As shown in fig. 1, an embodiment of the present invention provides a data stream arranging system. Wherein the data stream orchestration system may comprise a memory and a processor.

In detail, in one application example, the memory and the processor are electrically connected directly or indirectly to realize transmission or interaction of data. For example, electrical connection may be made to each other via one or more communication buses or signal lines. The memory may store at least one software functional module (computer program) that may exist in the form of software or firmware. The processor may be configured to execute the executable computer program stored in the memory, so as to implement the data stream arranging method applied to the oil and gas production system provided by the embodiment of the invention.

In detail, in one application example, the Memory may be, but is not limited to, a random access Memory (Random Access Memory, RAM), a Read Only Memory (ROM), a programmable Read Only Memory (Programmable Read-Only Memory, PROM), an erasable Read Only Memory (Erasable Programmable Read-Only Memory, EPROM), an electrically erasable Read Only Memory (Electric Erasable Programmable Read-Only Memory, EEPROM), or the like. The processor may be a general purpose processor including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), a System on Chip (SoC), etc.; but also Digital Signal Processors (DSPs), application Specific Integrated Circuits (ASICs), field Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

In detail, in an application example, the data stream orchestration system may be a server or a cluster of servers, etc. with data processing capabilities.

With reference to fig. 2, the embodiment of the invention further provides a data stream arranging method applied to the oil and gas production system, which can be applied to the data stream arranging system. The method steps defined by the flow related to the data flow arrangement method applied to the oil and gas production system can be realized by the data flow arrangement system. The specific flow shown in fig. 2 will be described in detail.

Step S110, equipment operation state data of oil gas production equipment in a target oil gas production system is obtained, and the equipment operation state data is subjected to feature mining by utilizing a feature mining unit included in a monitoring data matching network, so that equipment operation state features corresponding to the equipment operation state data are output.

In the embodiment of the invention, the data stream arrangement system can acquire the equipment operation state data of the oil gas production equipment in the target oil gas production system, and utilizes the feature mining unit included in the monitoring data matching network to perform feature mining on the equipment operation state data, and output the equipment operation state features corresponding to the equipment operation state data for representing the semantics of the equipment operation state data. The equipment operation state data is used for describing the operation process of the oil and gas production equipment, for example, the operation process of the oil and gas production equipment can be monitored through a sensor to form the equipment operation state data, or an operation log of the oil and gas production equipment can be extracted to serve as the equipment operation state data. The feature mining unit may be a convolutional neural network comprising one or more convolutional kernels.

Step S120, using a first data matching unit included in the monitoring data matching network, performing subsystem information matching on the device running state data according to the device running state feature, and outputting matching storage subsystem information corresponding to the device running state data in the target storage frame information.

In the embodiment of the invention, the data stream arranging system can utilize the first data matching unit included in the monitoring data matching network to match the subsystem information of the equipment running state data according to the equipment running state characteristics, and output the matched storage subsystem information corresponding to the equipment running state data in the target storage frame information. The target storage frame information comprises at least two storage subsystem information and a plurality of storage device information formed by subdividing the at least two storage subsystem information, each storage device information comprises at least one representative storage data segment, the storage subsystem information is used for reflecting one storage subsystem in a storage system of the equipment operation state data, each storage subsystem comprises at least two storage devices, each storage device corresponds to one storage device information, and at least one representative storage data segment included in the storage device information is used for representing historical equipment operation state data stored by the corresponding storage device. The first data matching unit may include a full connection subunit and a softmax subunit, and after the full connection subunit processes the operating state feature of the device, the softmax subunit is used to process the obtained full connection feature to obtain a probability distribution, where the probability distribution includes probability parameters corresponding to each storage subsystem information, and then one storage subsystem information with the largest probability parameter may be used as the matching storage subsystem information corresponding to the operating state data of the device.

And step S130, performing equipment information matching on the equipment operation state data according to the equipment operation state characteristics and the storage equipment characteristics of each piece of storage equipment information by using a second data matching unit included in the monitoring data matching network, and outputting matched storage equipment information corresponding to the equipment operation state data in the target storage frame information.

In the embodiment of the present invention, the data stream arranging system may utilize a second data matching unit included in the monitoring data matching network to perform device information matching on the device running state data according to the device running state feature and the storage device feature of each storage device information, and output matching storage device information corresponding to the device running state data in the target storage frame information. The storage device characteristics are determined according to representative storage data fragments included in corresponding storage device information, and the matched storage device information belongs to one storage device information of at least two storage device information corresponding to the matched storage subsystem information. The second data matching unit may include a dimension reduction subunit and a softmax subunit (i.e., constitute a data matching subunit described later), after the dimension reduction subunit performs dimension reduction processing on the device running state feature and the aggregate feature of the storage device feature of each storage device information, the softmax subunit is used to process the dimension reduced aggregate feature to obtain a probability distribution, where the probability distribution includes probability parameters corresponding to each storage device information, and then one storage device information with the largest probability parameter may be used as the matching storage device information corresponding to the device running state data.

And step S140, arranging the equipment operation state data based on the corresponding matched storage equipment information, so that the equipment operation state data is stored in the storage equipment corresponding to the matched storage equipment information.

In the embodiment of the invention, the data stream arrangement system can arrange the equipment running state data based on the corresponding matched storage equipment information, so that the equipment running state data is stored in the storage equipment corresponding to the matched storage equipment information. That is, for each oil gas production device in the target oil gas production system, after obtaining the matching storage device information corresponding to the device operation state data of each oil gas production device, the matching storage device information may be used as an arrangement basis of the device operation state data of each oil gas production device, and the device operation state data with the same matching storage device information may be distributed together and sent to the corresponding storage device for storage.

Based on the foregoing, on the one hand, because two-level data matching is performed, the accuracy of data matching is higher, on the other hand, because the storage device characteristics of the storage device information corresponding to the device running state data are determined by representing the storage data segments, and the representing storage data segments can represent the semantics expressed by the storage device information to a certain extent, the storage device characteristics corresponding to the storage device information and the device running state characteristics of the device running state data are determined based on the representing storage data segments of the storage device information, the device running state data are subjected to device information matching, the obtained matching storage device information to which the device running state data belongs can be more accurate, and the accuracy of data matching for the device running state data can be improved, so that the problem that the reliability of data stream arrangement in the prior art is relatively not high is solved.

For example, in the field of oil and gas production, production facilities primarily refer to facilities for extracting and processing crude oil and natural gas from subsurface oil and gas reservoirs, with the following being some common oil and gas production facilities:

drilling equipment: including drills, drill rods, drill bits, etc., for drilling boreholes in the subsurface for accessing hydrocarbon reservoirs; oil extraction equipment: the system comprises a pump rod, a pumping unit, an artificial lifting system and the like, and is used for pumping crude oil from an oil well shaft to the ground; separation equipment: such as separators, settling tanks, and rotating separators, for separating crude oil extracted from a wellhead from natural gas; storage device: the device comprises an oil tank, a gas storage tank and the like, and is used for storing extracted crude oil and natural gas; processing equipment: such as oil and gas treatment plants, desulfurization units, dehydration equipment, etc., for processing and treating crude oil and natural gas, improving its quality and adapting to market demands; conveying equipment: including pipes, valves, pump stations, etc., for transporting crude oil and natural gas to processing plants, storage facilities, or direct supply markets; compression equipment: the natural gas compressor comprises a compressor, a booster pump and the like, and is used for compressing natural gas into a high-pressure state, so that the natural gas is convenient to store, transport or further process; logging equipment: such as logging instruments and sensors, for obtaining information about subsurface hydrocarbon reservoirs, such as wellbore diameter, permeability, fluid properties, etc.; water treatment equipment: to water source purification, sedimentation tanks, filters, etc., for treating the water content produced from the well and bringing it to the environmental regulations; explosion-proof equipment: such as explosion-proof electrical equipment, explosion-proof lighting and explosion-proof ventilation systems, for ensuring safe operation in flammable gas environments; water injection equipment: comprises a water injection pump, a water injection well and the like, for injecting water into an oil well to maintain oil pressure, improve oil recovery efficiency and extend oil field life; heating equipment: such as furnaces, heaters, and heat exchangers, for heating crude oil, natural gas, or other fluids to improve their flowability or to meet specific process requirements; a pressure vessel: for storing and transporting high pressure gases or liquids, such as compressed air tanks, liquefied Petroleum Gas (LPG) storage tanks, etc.; gas detection device: the system comprises a gas detector and a monitoring system, which are used for monitoring and detecting harmful gases and explosion risks in oil wells and production equipment; leak detection apparatus: such as leak detectors, flow meters, etc., for timely detection and alarm processing of possible oil and gas leakage conditions.

For example, in oil and gas production, there is a close correlation and synergy between the devices, the following are some of the common devices:

drilling equipment and logging equipment: the well equipment is used to open a well, and the well equipment provides information about the subsurface reservoir; the logging equipment can measure in the drilling process, and help to determine parameters such as thickness of hydrocarbon reservoir, hydrocarbon-bearing property, permeability and the like, so as to guide subsequent production decisions;

oil extraction equipment and water injection equipment: the oil extraction equipment is used for extracting the crude oil from the oil well to the ground, and the water injection equipment is responsible for injecting water into the oil well to maintain the oil pressure and increase the oil extraction efficiency; the two devices work together to maintain the pressure balance of the oil field by controlling the oil extraction and water injection processes, so that the service life of the oil field is prolonged;

separation apparatus and storage apparatus: the separation equipment is used for separating crude oil extracted from a wellhead from natural gas for further processing and transportation; the separated crude oil and natural gas can be stored by storage equipment such as oil tanks, gas storage tanks and the like, and waiting for subsequent processing, selling or transportation;

processing equipment and compression equipment: the processing equipment is used for processing and treating crude oil and natural gas so as to meet market demands and improve the quality and the value of the crude oil and the natural gas; during the processing, the natural gas may need to be compressed into a high pressure state by using compression equipment, which is convenient for storage, transportation or further treatment;

Conveying equipment and storage equipment: conveying equipment such as pipes, valves, pump stations, etc. for conveying crude oil and natural gas from a production site to a processing plant, storage facility, or direct supply market; the storage device is used for temporarily storing crude oil and natural gas so as to carry out reasonable dispatching and distribution according to market demands.

For example, the plant operational status data for each oil and gas production plant may be:

drilling equipment: 2023-09-30; the operator: john Smith; drill bit type: PDC; well depth (feet): 10,000; drilling rate (feet per hour): 50; mud density (pounds per gallon): 12; cyclic Pressure (PSI): 5,000;

oil extraction equipment: 2023-09-30; production well number: WELL001; well production (barrels per day): 500; well production (barrels/day): 200; deposit content: 0.5%; electric pump run time (hours): 8, 8; current load (amperes): 75;

separation equipment: 2023-09-30; feed flow (cubic meters per hour): 100; oil phase outlet flow (cubic meters per hour): 80; aqueous phase outlet flow (cubic meters per hour): 20, a step of; separation temperature (degrees celsius): 70; separation pressure (bar): 2;

Storage device: 2023-09-30; number of storage tank: TANK001; the storage substance: crude oil; tank capacity (cubic meters): 1000; current tank level (percent): 80%; tank temperature (degrees celsius): 25, a step of selecting a specific type of material;

processing equipment: 2023-09-30; the processing device comprises: a distillation column; feed flow (cubic meters per hour): 200; product outlet flow (cubic meters per hour): 150; temperature (degrees celsius): 120; pressure (bar): 5, a step of;

conveying equipment: 2023-09-30; conveyor belt numbering: CONVEYOR001; material type: crude oil; transport speed (meters/minute): 0.5; load weight (ton): 10; run time (hours): 8, 8;

compression equipment: 2023-09-30; compressor model: COMPRESOR 001; intake temperature (degrees celsius): 25, a step of selecting a specific type of material; temperature of outgassing (degrees celsius): 80; intake pressure (bar): 1, a step of; gas outlet pressure (bar): 10;

logging equipment: 2023-09-30; logging tools: resistivity tools; depth of measurement (feet): 8000; resistivity measurements (ohm-meter): 2; mud density (pounds per gallon): 10; temperature (degrees celsius): 40, a step of performing a;

water treatment equipment: 2023-09-30; a water treatment device: a reverse osmosis system; water inflow (cubic meter/hour): 50; effluent flow (cubic meter/hour): 40, a step of performing a; treatment efficiency (percent): 80%; operating pressure (bar): 10;

Explosion-proof equipment: 2023-09-30; device type: an explosion-proof motor; motor power (kw): 50; explosion protection rating: ex d IIB T4 Gb; operating state: normal; temperature (degrees celsius): 60;

water injection equipment: 2023-09-30; water injection well numbering: INJECTION001; injection flow rate (cubic meters per hour): 100; injection pressure (bar): 20, a step of; and (3) water quality monitoring: pH and dissolved oxygen content; the operator: jane Doe;

heating equipment: 2023-09-30; heater model: HEATER001; heating medium: steam; feed temperature (degrees celsius): 20, a step of; discharge temperature (degrees celsius): 80; heating power (kw): 100;

a pressure vessel: 2023-09-30; container number: VESSEL001; type of container: a storage tank; maximum design pressure (bar): 10; current pressure (bar): 8, 8; temperature (degrees celsius): 30;

gas detection device: 2023-09-30; the model of the detector: gasetector 001; monitoring gas type: methane, carbon monoxide; gas concentration (percent): methane: 1%, carbon monoxide: 0.5%; alarm threshold: methane: 5%, carbon monoxide: 10%;

leak detection apparatus: 2023-09-30; leak monitor model: leakdetect 001; monitoring position: a pipe joint; current state: normal; alarm threshold: the flow exceeds 10 cubic meters per hour.

The foregoing is a simple example of equipment operational status data, which is illustrated in greater detail using an oil and gas production facility as a drilling facility:

2023-09-30; 08:00-12:00 time; drilling equipment: the operator: john Smith; workshop number: DRILL001; well name: wellA-01; well depth (feet): 10,000; drill bit type: PDC; drill bit size: 8.5 inches; drilling rate (feet per hour): 50; drilling fluid type: water-based mud; mud density (pounds per gallon): 12.5; cyclic Pressure (PSI): 4,800; circulation flow (GPM): 2,500; cycling temperature (degrees celsius): 35; bottom hole Pressure (PSI): 6,500; well killing displacement (barrel): 200; well-killing pump truck: pumpTruck001; drilling time (hours): 4, a step of; drilling records: drilling machine tripping time: 08:00; surface casing down time: 08:30; contact to the formation: shale; description of the stratum: compact and high in oil content; drilling speed (hours/meter): 2; total drilling time (hours): 4, a step of; drilling fluid circulation rate (percent): 95%; drilling fluid performance detection results: the density is normal, and the viscosity is normal; drilling tool usage record: drill pipe type: HWDP (Heavy Weight Drill Pipe); number of drill pipes: 10 roots; drill pipe length (feet): 30; drill collar type: a Mud Motor; drill collar dimensions: 6.25 inches; drill collar rotation speed (RPM): 80; drill collar power (horsepower): 500; drill collar torque (ft-lb): 10,000; stratum sampling record: core extractor model: coreBarrel001; core diameter (inches): 4, a step of; core length (feet) of core coring: 10; coring start time: 09:00; coring end time: 09:30; core sample description: shale samples, wet, have obvious oil marks; drilling environment parameter record: atmospheric Pressure (PSI): 14.7; ambient temperature (degrees celsius): 30; wind speed (feet/second): 5, a step of; rainfall (mm): 0; recording safety accidents: no safety accident occurs;

2023-09-30; time is 12:00-16:00; drilling equipment: the operator: john Smith; workshop number: DRILL001; well name: wellA-01; drill bit type: PDC; drill bit size: 8.5 inches; drilling rate (feet per hour): 45; drilling fluid type: water-based mud; mud density (pounds per gallon): 12.2; cyclic Pressure (PSI): 4,600; circulation flow (GPM): 2,400; cycling temperature (degrees celsius): 34; bottom hole Pressure (PSI): 6,300; well killing displacement (barrel): 180; well-killing pump truck: pumpTruck001; drilling time (hours): 4, a step of; drilling records: drilling machine tripping time: 12:00; drill pipe inspection time: 12:30; drill rod wear conditions: normal, no cracking or deformation; drilling fluid performance detection results: the density is normal, and the viscosity is normal; drill floor maintenance record: cleaning rock debris and replacing a dust cover; drilling tool usage record: drill pipe type: HWDP (Heavy Weight Drill Pipe); number of drill pipes: 10 roots; drill pipe length (feet): 30; drill collar type: a Mud Motor; drill collar dimensions: 6.25 inches; drill collar rotation speed (RPM): 75; drill collar power (horsepower): 480. Drill collar torque (ft-lb): 9,500; stratum sampling record: core extractor model: coreBarrel001; core diameter (inches): 4, a step of; core length (feet) of core coring: 10; coring start time: 13:00; coring end time: 13:30; core sample description: shale samples, wet, have small amounts of oil marks; drilling environment parameter record: atmospheric Pressure (PSI): 14.5; ambient temperature (degrees celsius): 32; wind speed (feet/second): 8, 8; rainfall (mm): 0; recording safety accidents: no safety accident occurs;

2023-09-30; time is 16:00-20:00; drilling equipment: the operator: john Smith; workshop number: DRILL001; well name: wellA-01; drill bit type: PDC; drill bit size: 8.5 inches; drilling rate (feet per hour): 47; drilling fluid type: water-based mud; mud density (pounds per gallon): 12.0; cyclic Pressure (PSI): 4,400; circulation flow (GPM): 2,300; cycling temperature (degrees celsius): 33; bottom hole Pressure (PSI): 6,100; well killing displacement (barrel): 170, a step of; well-killing pump truck: pumpTruck001; drilling time (hours): 4, a step of; drilling records: drilling machine tripping time: 16:00; description of the stratum: the shale layer transitions to the sandstone layer; drilling speed (hours/meter): 1.5; total drilling time (hours): 8, 8; drilling fluid circulation rate (percent): 90%; drilling fluid performance detection results: the density is normal, and the viscosity is normal; drilling tool usage record: drill pipe type: HWDP (Heavy Weight Drill Pipe); number of drill pipes: 10 roots; drill pipe length (feet): 30; drill collar type: a Mud Motor; drill collar dimensions: 6.25 inches; drill collar rotation speed (RPM): 70; drill collar power (horsepower): 460; drill collar torque (ft-lb): 9,000; stratum sampling record: core extractor model: coreBarrel001; core diameter (inches): 4, a step of; core length (feet) of core coring: 10; coring start time: 17:00; coring end time: 17:30; core sample description: the sandstone sample is dried and has no oil stain; drilling environment parameter record: atmospheric Pressure (PSI): 14.4; ambient temperature (degrees celsius): 31; wind speed (feet/second): 7, preparing a base material; rainfall (mm): 0; recording safety accidents: no safety accident occurs.

Following the above example, the storage system for the plant operational status data of each of the hydrocarbon production plants in the target hydrocarbon production system may be divided into three storage subsystems, storage subsystem a, storage subsystem B, and storage subsystem C. The storage subsystem A is used for storing abnormal-free equipment operation state data, the storage subsystem A comprises a storage equipment a and a storage equipment b, the storage equipment a is used for storing abnormal-free equipment operation state data of each oil gas production equipment which is not related to other oil gas production equipment, and the storage equipment b is used for storing abnormal-free equipment operation state data of each oil gas production equipment which is related to other oil gas production equipment. The storage subsystem B is used for storing abnormal equipment operation state data belonging to internal abnormality, the storage subsystem B comprises a storage equipment c and a storage equipment d, the storage equipment c is used for storing abnormal equipment operation state data belonging to internal abnormality and having no association relation with other oil gas production equipment, and the storage equipment d is used for storing abnormal equipment operation state data belonging to internal abnormality and having association relation with other oil gas production equipment. The storage subsystem C is used for storing abnormal equipment operation state data belonging to external abnormality, the storage subsystem C comprises a storage equipment e and a storage equipment f, the storage equipment e is used for storing abnormal equipment operation state data belonging to external abnormality and having no association relation with other oil gas production equipment, and the storage equipment f is used for storing abnormal equipment operation state data belonging to external abnormality and having association relation with other oil gas production equipment.

Next, the internal abnormality and the external abnormality of each oil and gas production facility will be described by way of the above example:

drilling equipment: external anomalies: environmental anomalies such as geologic condition changes, weather effects, wellhead accidents, etc.; internal anomalies: abnormal conditions related to the equipment such as drilling tool faults, drilling fluid runaway, abnormal circulation systems, borehole collapse and the like;

oil extraction equipment: external anomalies: such as oil well sand plugging, sediment plugging, wellhead corrosion, etc., associated with the oil producing environment; internal anomalies: anomalies related to the oil extraction equipment such as pump rod fracture, motor faults, water injection anomalies, pressure fluctuations and the like;

separation equipment: external anomalies: anomalies associated with feed conditions and operating environment such as feed concentration variations, crude oil composition fluctuations, operating errors, etc.; internal anomalies: abnormal conditions related to the separation equipment such as equipment blockage, degradation of separation effect, sensor failure, etc.;

storage device: external anomalies: abnormalities associated with environmental conditions such as temperature changes, external pressure fluctuations, disaster events, etc.; internal anomalies: including anomalies associated with the storage device itself, such as tank pressure anomalies, liquid level anomalies, leaks, etc.;

Processing equipment: external anomalies: anomalies associated with external conditions and input materials such as raw material quality variations, power supply instability, process adjustments, etc.; internal anomalies: the method relates to equipment faults, abnormal processing effects, control system faults and other faults related to processing equipment;

conveying equipment: external anomalies: anomalies associated with the transport environment and conditions, such as pipe damage, climate effects, transport medium changes, etc.; internal anomalies: anomalies associated with the delivery device itself, including pipe blockage, valve failure, flow anomalies, etc.;

compression equipment: external anomalies: abnormality related to external conditions such as abnormal power supply, change in ambient temperature, unsmooth ventilation, and the like; internal anomalies: abnormal conditions related to the compression equipment itself, such as compressor failure, pressure fluctuations, vibration anomalies, etc.;

logging equipment: external anomalies: anomalies associated with logging environment and geological conditions, such as logging tool damage, borehole instability, formation anomalies, etc.; internal anomalies: abnormal conditions related to logging equipment such as logging instrument faults, sensor failures, abnormal data acquisition and the like;

water treatment equipment: external anomalies: such as water quality changes of inflow water, pollution sudden increase in the treatment process, unstable power supply and other anomalies related to external conditions and input water quality; internal anomalies: abnormal conditions related to the water treatment equipment such as equipment blockage, reduced treatment effect, control system failure and the like;

Explosion-proof equipment: external anomalies: abnormalities related to the surrounding environment such as explosion threats, increases in the concentration of ambient gases, etc.; internal anomalies: relates to failure of an explosion-proof device and loss of a safety valve;

water injection equipment: external anomalies: such as water injection quality change, water injection well plugging, water injection pipeline leakage and other anomalies related to external conditions and environments; internal anomalies: the water injection device comprises water injection pump faults, abnormal water injection pressure, abnormal water injection flow and other anomalies related to the water injection device;

heating equipment: external anomalies: anomalies associated with external conditions and input media, such as energy supply instability, media temperature changes, heating media mass changes, etc.; internal anomalies: anomalies related to the heating device itself, such as heating element faults, temperature control deviations, heating medium leaks, etc.;

a pressure vessel: external anomalies: anomalies associated with external conditions and environments such as ambient temperature changes, climate effects, external pressure fluctuations, etc.; internal anomalies: abnormalities associated with the pressure vessel itself, such as pressure vessel leaks, pressure anomaly changes, structural damage, etc.;

gas detection device: external anomalies: anomalies associated with external conditions and operations, such as increased ambient gas concentration, operating errors, calibration inaccuracies, etc.; internal anomalies: abnormal conditions related to the gas detection device such as sensor faults, abnormal signal processing, abnormal data acquisition and the like;

Leak detection apparatus: external anomalies: anomalies associated with external conditions and operations such as increased ambient gas concentration, operating errors, calibration inaccuracies, etc.; internal anomalies: including anomalies associated with the leak detection apparatus itself, such as leak detection sensor faults, signal processing anomalies, data acquisition anomalies, and the like.

It will be appreciated that the above example is merely illustrative, and that in actual practice, the division of the storage system subsystems and storage devices may be configured based on other requirements.

In detail, in an application example, the above step S110 may include:

determining representative data segments from the device operation state data to form at least one state representative data segment included in the device operation state data, where the device operation state data may be text data, and based on this, determining representative data segments may refer to determining keywords from the text data, where the determining of keywords may use related existing technologies, for example, determining keywords based on statistical features, for example, extracting words with higher Frequency from a corpus database as candidate keywords by using statistical methods and algorithms, for example, using TF-IDF (Term Frequency-Inverse Document Frequency) algorithm or word Frequency-based method to calculate importance of words, and then using the candidate keywords appearing in the device operation state data as state representative data segments to obtain at least one state representative data segment;

The characteristic mining unit is used for respectively performing characteristic mining on all state representing data fragments included in the equipment operation state data and outputting state representing data fragment characteristics corresponding to all state representing data fragments; for example, as described above, the state representing data segment may be a text word, so word embedding processing may be performed on the text word to obtain a corresponding word embedding vector, which is used as a corresponding state representing data segment feature; or, the embedded vectors of the text distribution positions of the state representative data segments can be overlapped on the basis of the word embedded vectors to obtain the corresponding state representative data segment characteristics;

aggregating the output states to represent the data segment characteristics to form the equipment operation state characteristics corresponding to the equipment operation state data; for example, the state representing data segment features may be spliced or cascaded to obtain the device operation state feature corresponding to the device operation state data.

For example, the process of obtaining the state representing data segment characteristics corresponding to the state representing data segment through the word embedding model may be as follows:

Assuming that the state representing data segment includes "stratum description", "shale layer", "transition to", "sandstone layer", "stratum description", the text distribution position corresponding to "stratum description" is "00", the text distribution position corresponding to "shale layer" is "01", "the text distribution position corresponding to" transition to "is" 02", and the text distribution position corresponding to" sandstone layer "is" 04";

"formation description": word embedding vector: -0.12, 0.45, 0.78, -0.34], the corresponding text distribution position is "00", the corresponding word embedding vector: [0.1, -0.2, 0.3,..0.5 ], the superimposed state representing the data fragment characterized by, [ -0.02, 0.25, 1.08,..0.16 ];

"shale layer": word embedding vector: [0.67, -0.56, 0.23,..0.89 ], the corresponding text distribution position is "01", the corresponding word embedding vector: -0.3, 0.5, -0.1, & gt, 0.2] the superimposed state representing a data fragment characterized by [0.37, -0.06, 0.13, & gt, 1.09];

transition to: word embedding vector: [0.43, 0.21, -0.78,..0.12 ], the corresponding text distribution position is "02", the corresponding word embedding vector: [0.2, 0.4, -0.3, -0.1], the superimposed state representing a data segment characterized by [0.63, 0.61, -1.08, 0.02];

"sandstone layer": word embedding vector: [ -0.98, -0.01, 0.76, ], 0.34], the corresponding text distribution position is "04", the corresponding word embedding vector: the superimposed state is characterized by data fragments [ (0.5, -0.2, 0.3 ], 0.4 ]).

In detail, in an application example, the above step S120 may include:

determining the at least two pieces of storage subsystem information included in the target storage frame information, wherein the storage system comprises a storage subsystem A, a storage subsystem B and a storage subsystem C, and is used for storing abnormal-free equipment operation state data, abnormal-internal abnormal equipment operation state data and external abnormal equipment operation state data respectively;

using a first data matching unit included in the monitoring data matching network, according to the equipment operation state characteristics and the at least two pieces of storage subsystem information, performing data matching on the equipment operation state data by the at least two pieces of storage subsystem information, and outputting a possibility parameter that the equipment operation state data is matched with each piece of storage subsystem information; that is, after the full connection sub-unit in the first data matching unit performs full connection processing on the device running state feature to obtain the full connection feature, the full connection feature is processed by using the softmax sub-unit in the first data matching unit to obtain the possibility parameter of each storage sub-system information, that is, the possibility parameter 1 of the device running state data belonging to the device running state data without abnormality, the possibility parameter 2 of the device running state data with abnormality and internal abnormality, and the possibility parameter 3 of the device running state data with abnormality and external abnormality;

Marking storage subsystem information corresponding to the possibility parameter with the maximum value to form matching storage subsystem information corresponding to the equipment running state data; illustratively, for example, the likelihood parameter 1, the likelihood parameter 2, and the likelihood parameter 3 are respectively 0.3, 0.7, and 0.1, and thus the storage subsystem B corresponding to the likelihood parameter 2 may be used as the matching storage subsystem information.

In detail, in an application example, the second data matching unit may include a feature aggregation subunit and a data matching subunit, based on which, the step of using the second data matching unit included in the monitoring data matching network to match the device running state data with the device running state data according to the device running state feature and the storage device feature of each storage device information, and outputting matched storage device information corresponding to the device running state data in the target storage frame information may include:

the feature aggregation subunit is utilized to aggregate the equipment running state features and the storage equipment features corresponding to the storage equipment information to form corresponding aggregation features, so that the aggregation features comprise semantic information of the equipment running state features and the storage equipment features corresponding to the storage equipment information;

Using the data matching subunit to match the equipment operation state data with the equipment information according to the aggregation characteristics, and outputting corresponding matched storage equipment information of the equipment operation state data in the target storage frame information; the data matching subunit may pool the aggregate features to implement dimension reduction, process the dimension reduced features through a softmax function to obtain a likelihood parameter of each storage device information, and use the storage device information corresponding to the likelihood parameter having the maximum value as the matching storage device information; based on this, since the storage device characteristics corresponding to each of the storage device information characterize the semantics of the already stored historical device operation state data, for example, the semantics of the historical device operation state data already stored by the storage device a are: abnormal-free equipment operation state data of each oil gas production equipment which is not associated with other oil gas production equipment, and the semantics of the historical equipment operation state data stored by the storage equipment b are as follows: the semantics of the historical equipment operation state data stored by the storage equipment c are as follows: the device operation state data of each oil gas production device which is abnormal and belongs to internal abnormality and has no association relation with other oil gas production devices, and the semantics of the historical device operation state data stored by the storage device d are as follows: the device operation state data of each oil gas production device which has abnormality and belongs to internal abnormality and has association relation with other oil gas production devices, and the semantics of the historical device operation state data stored by the storage device e are as follows: the device operation state data of each oil gas production device which is abnormal and belongs to external abnormality and has no association relation with other oil gas production devices is stored, and the semantics of the stored historical device operation state data of the storage device f are as follows: the equipment operation state data of each oil gas production equipment which is abnormal and belongs to external abnormality and has association relation with other oil gas production equipment. Therefore, by aggregating the storage device characteristics corresponding to the storage device information, the semantics of various classifications can be enhanced, and the reliability of the probability parameters of the obtained storage device information is higher.

Next to the above example, the storage subsystem B is used as the matching storage subsystem information, and the storage subsystem B includes a storage device c for storing the device operation state data of each of the oil and gas production devices having abnormality and belonging to internal abnormality and having no association with other oil and gas production devices, and a storage device d for storing the device operation state data of each of the oil and gas production devices having abnormality and belonging to internal abnormality and having association with other oil and gas production devices. Thus, after determining, based on step S120, that the equipment operation state data belongs to equipment operation state data having an abnormality and belonging to an internal abnormality, it may be further determined whether or not the equipment operation state data belongs to an oil and gas production equipment having no association with other oil and gas production equipment. If the probability parameter of the equipment operation state data of the oil and gas production equipment which has no association relation with other oil and gas production equipment is 0.2, the probability parameter of the equipment operation state data of the oil and gas production equipment which has association relation with other oil and gas production equipment is 0.7; alternatively, in other embodiments, 3*2 =6 likelihood parameters may be obtained, so that it may be determined that the device operation state data belongs to abnormal and internal abnormal device operation state data of each oil and gas production device having an association relationship with other oil and gas production devices, and thus the corresponding matching storage device information is the storage device d.

In detail, in an application example, the step of using the feature aggregation subunit to aggregate the device running state feature and the storage device feature corresponding to each piece of storage device information to form a corresponding aggregate feature may include:

determining a first modal linear mapping parameter for performing first modal linear mapping, wherein the first modal linear mapping parameter can be formed in the updating and optimizing process of the neural network;

determining a second modality linear mapping parameter for performing a second modality linear mapping, wherein the second modality linear mapping parameter can be formed in an updating and optimizing process of the neural network;

performing second-mode linear mapping on the equipment running state characteristics and the storage equipment characteristics of each piece of storage equipment information according to the second-mode linear mapping parameters by using the characteristic aggregation subunit to form corresponding second-mode linear mapping characteristics, wherein the second-mode linear mapping and the first-mode linear mapping can be two different linear mappings, such as superposition mapping, multiplication mapping and the like;

Utilizing the characteristic aggregation subunit to aggregate the first modality linear mapping characteristic and the second modality linear mapping characteristic to form a corresponding aggregation characteristic; for example, the first modality linear mapping feature and the second modality linear mapping feature may be superimposed to achieve aggregation.

In detail, in an application example, the step of performing, by using the feature aggregation subunit, first-modality linear mapping on the device running state feature and the storage device feature of each piece of storage device information according to the first-modality linear mapping parameter to form a corresponding first-modality linear mapping feature may include:

splicing the equipment running state characteristics and the storage equipment characteristics of the storage equipment information to obtain corresponding first aggregation characteristics;

and based on a first weight parameter included in the first modal linear mapping parameter, weighting the first aggregation feature to obtain a corresponding first weighted aggregation feature, and based on a first bias parameter included in the first modal linear mapping parameter, superposing the first weighted aggregation feature to obtain a corresponding first modal linear mapping feature.

In detail, in an application example, the step of performing, by using the feature aggregation subunit, second-modality linear mapping on the device running state feature and the storage device feature of each piece of storage device information according to the second-modality linear mapping parameter to form a corresponding second-modality linear mapping feature may include:

splicing the storage device characteristics of the storage device information to form corresponding second aggregation characteristics, multiplying the equipment operation state characteristics, second weight parameters included in the second modal linear mapping parameters and the second aggregation characteristics to obtain corresponding second weighted aggregation characteristics, and superposing the second weighted aggregation characteristics based on second bias parameters included in the second modal linear mapping parameters to obtain corresponding second modal linear mapping characteristics;

if the dimension of the running state feature of the device is 1*n, the dimension of the second aggregation feature is m×1, the dimension of the second weight parameter is n×m×y, and the dimension of the obtained second weighted aggregation feature is 1*y.

In detail, in another application example, the step of using the feature aggregation subunit to aggregate the device running state feature and the storage device feature corresponding to each piece of storage device information to form a corresponding aggregate feature may include:

Determining a first modality linear mapping parameter for performing a first modality linear mapping; and performing first-modality linear mapping on the device running state characteristics and the storage device characteristics of each piece of storage device information according to the first-modality linear mapping parameters by using the characteristic aggregation subunit to form corresponding first-modality linear mapping characteristics, as described in the foregoing related description; and obtaining corresponding aggregation features based on the first modality linear mapping features; illustratively, the first modality linear mapping feature may be directly taken as a corresponding aggregation feature;

determining a second modality linear mapping parameter for performing a second modality linear mapping; and performing second-modality linear mapping on the device running state characteristics and the storage device characteristics of each piece of storage device information according to the second-modality linear mapping parameters by using the characteristic aggregation subunit to form corresponding second-modality linear mapping characteristics, as described in the foregoing related description; and obtaining a corresponding aggregation feature based on the second modality linear mapping feature; for example, the second modality linear mapping feature may be directly treated as a corresponding aggregation feature.

In detail, in an application example, before the step of performing subsystem information matching on the equipment operation state data according to the equipment operation state feature by using the first data matching unit included in the monitoring data matching network and outputting the matched storage subsystem information corresponding to the equipment operation state data in the target storage frame information, the data stream arrangement method applied to the oil and gas production system may further include:

determining representative storage data fragments included in each piece of storage equipment information in the target storage frame information, namely determining keywords, as described in the previous correlation;

for each piece of storage equipment information, performing feature mining on each representative storage data segment included in the storage equipment information to form a representative storage data segment feature corresponding to each representative storage data segment, and if the representative storage data segment feature is embedded through a word embedding model, superposing vectors representing positions;

performing feature aggregation on the storage device information including the features of the representative storage data fragments corresponding to the representative storage data fragments to form storage device features corresponding to the storage device information; that is, for each piece of storage device information, feature aggregation is performed on the representative storage data segment features corresponding to the representative storage data segments included in the storage device information, so as to form storage device features corresponding to the storage device information.

In detail, in an application example, the step of performing feature aggregation on the storage device information including the representative storage data segment feature corresponding to the representative storage data segment to form the storage device feature corresponding to each storage device information may include:

for each piece of storage equipment information, when the number of fragments of the representative storage data fragments included in the storage equipment information is greater than or equal to the predetermined reference fragment number, carrying out averaging processing on the characteristic of the representative storage data fragments corresponding to the representative storage data fragments included in the storage equipment information to form the storage equipment characteristic corresponding to the storage equipment information; the specific number of the reference fragments is not limited, and can be configured according to actual requirements, such as 3, 5, 9 and the like;

In detail, in an application example, before the step of obtaining the equipment operation state data of the oil gas production equipment in the target oil gas production system and using the feature mining unit included in the monitoring data matching network to perform feature mining on the equipment operation state data and output the equipment operation state feature corresponding to the equipment operation state data, the data stream arrangement method applied to the oil gas production system may further include:

the feature mining unit is used for feature mining of the training equipment operation state data and outputting training equipment operation state features corresponding to the training equipment operation state data, and the feature mining unit is used for feature mining of the training equipment operation state data and outputting training equipment operation state features corresponding to the training equipment operation state data as described above;

using the first data matching unit to match subsystem information of the training equipment operation state data according to the training equipment operation state characteristics, and outputting corresponding subsystem information matching data, wherein the subsystem information matching data is used for reflecting matched storage subsystem information estimated by the training equipment operation state data, as described above;

Using the second data matching unit to perform device information matching on the training device running state data according to the training device running state characteristics and the storage device characteristics of each piece of storage device information, and outputting corresponding device information matching data, where the device information matching data is used to reflect the matched storage device information estimated by the training device running state data, as described above;

and adjusting network parameters of the monitoring data matching network based on the subsystem information distinction and the storage device information distinction to form a trained monitoring data matching network, e.g. calculating an error based on the subsystem information distinction and the storage device information distinction, and adjusting the network parameters of the monitoring data matching network based on the error.

In detail, in an application example, the step of adjusting the network parameters of the monitoring data matching network to form a trained monitoring data matching network based on the subsystem information difference and the storage device information difference may include:

Determining a first error index calculation rule corresponding to the first data matching unit, a second error index calculation rule corresponding to the second data matching unit and a dependency error index calculation rule, wherein the first error index calculation rule and the second error index calculation rule can be Cross Entropy (Cross-Entropy), a measurement method for measuring the difference between probability distributions, the dependency error index calculation rule is used for restricting storage device information reflected by the device information matching data from being subordinate to storage subsystem information reflected by the subsystem information matching data, for example, determining the difference between a value calculated based on the second error index calculation rule and a value calculated based on the first error index calculation rule, comparing the sum value of the difference and a preset parameter with a target value, taking a larger value of the sum value as a calculation value of the dependency error index calculation rule, and the preset parameter and the target value are not limited, such as 0.23 and 0 respectively; that is, if the value calculated by the second error index calculation rule is larger than the value calculated based on the first error index calculation rule, the calculated target error index is larger, and if the value calculated by the second error index calculation rule is larger than the value calculated based on the first error index calculation rule, the calculated target error index is smaller, that is, the error emphasis on the storage device hierarchy is realized;

Determining a target error index calculation rule of the monitored data matching network according to the first error index calculation rule, the second error index calculation rule and the dependency error index calculation rule, wherein the target error index calculation rule can be a sum or weighted sum of calculated values of the first error index calculation rule, the second error index calculation rule and the dependency error index calculation rule, so as to obtain a target error index;

according to the target error index, adjusting network parameters of the monitoring data matching network to form a trained monitoring data matching network; for example, the network parameters may be adjusted in a direction that reduces the target error indicator.

For example, the real storage subsystem information is (storage subsystem A:1, storage subsystem B:0, storage subsystem C: 0), the subsystem information match data is (storage subsystem A:0.7, storage subsystem B:0.2, storage subsystem C: 0.3), the real storage device information is (storage device a:1, storage device B:0, storage device C:0, storage device d:0, storage device e:0, storage device f: 0), the device information match data is (storage device a:0.8, storage device B:0.4, storage device C:0.1, storage device d:0.1, storage device e:0.2, storage device f: 0.3); based on this, the calculated value corresponding to the first error index calculation rule is 0.3567, the calculated value corresponding to the second error index calculation rule is 0.223, and the calculated value corresponding to the dependency error index calculation rule is max (0.223-0.3567+0.23, 0) =0.0963, and thus, the target error index is 0.3567+0.2231+0.0963= 0.6761.

With reference to fig. 3, an embodiment of the present invention further provides a data flow arrangement device applied to an oil and gas production system, which can be applied to the data flow arrangement system. Wherein, the data flow arrangement device applied to the oil gas production system can comprise:

the data feature mining module is used for acquiring equipment operation state data of oil gas production equipment in a target oil gas production system, performing feature mining on the equipment operation state data by utilizing a feature mining unit included in a monitoring data matching network, and outputting equipment operation state features corresponding to the equipment operation state data, wherein the equipment operation state data is used for describing an operation process of the oil gas production equipment;

the subsystem information matching module is used for matching subsystem information of the equipment operation state data according to the equipment operation state characteristics by utilizing a first data matching unit included in the monitoring data matching network, outputting corresponding matched storage subsystem information of the equipment operation state data in target storage frame information, wherein the target storage frame information comprises at least two storage subsystem information and a plurality of storage equipment information formed by subdividing the at least two storage subsystem information, each storage equipment information comprises at least one representative storage data segment, each storage subsystem information is used for reflecting one storage subsystem in a storage system of the equipment operation state data, each storage subsystem comprises at least two storage equipment, each storage equipment corresponds to one storage equipment information, and at least one representative storage data segment included in the storage equipment information is used for representing historical equipment operation state data stored by the corresponding storage equipment;

The device information matching module is used for matching the device operation state data with the device information according to the device operation state characteristics and the storage device characteristics of the storage device information by utilizing the second data matching unit included in the monitoring data matching network, outputting the matched storage device information corresponding to the device operation state data in the target storage frame information, wherein the storage device characteristics are determined according to the representative storage data fragments included in the corresponding storage device information, and the matched storage device information belongs to one storage device information of at least two storage device information corresponding to the matched storage subsystem information;

and the state data arrangement module is used for arranging the equipment operation state data based on the corresponding matched storage equipment information so that the equipment operation state data is stored in the storage equipment corresponding to the matched storage equipment information.

In summary, the data stream arrangement method and system applied to the oil and gas production system provided by the invention can perform feature mining on the equipment operation state data and output the equipment operation state features corresponding to the equipment operation state data; according to the equipment running state characteristics, subsystem information matching is carried out on the equipment running state data, and corresponding matching storage subsystem information of the equipment running state data in the target storage frame information is output; according to the equipment running state characteristics and the storage equipment characteristics of each piece of storage equipment information, equipment running state data are subjected to equipment information matching, and corresponding matched storage equipment information of the equipment running state data in the target storage frame information is output; and arranging the equipment running state data based on the corresponding matched storage equipment information. Based on the foregoing, on the one hand, because two-level data matching is performed, the accuracy of data matching is higher, on the other hand, because the storage device characteristics of the storage device information corresponding to the device running state data are determined by representing the storage data segments, and the representing storage data segments can represent the semantics expressed by the storage device information to a certain extent, the storage device characteristics corresponding to the storage device information and the device running state characteristics of the device running state data are determined based on the representing storage data segments of the storage device information, the device running state data are subjected to device information matching, the obtained matching storage device information to which the device running state data belongs can be more accurate, and the accuracy of data matching for the device running state data can be improved, so that the problem that the reliability of data stream arrangement in the prior art is relatively not high is solved.

The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, 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 data stream scheduling method for an oil and gas production system, comprising:

arranging the equipment operation state data based on the corresponding matched storage equipment information, so that the equipment operation state data is stored in a storage equipment corresponding to the matched storage equipment information;

the second data matching unit comprises a feature aggregation subunit and a data matching subunit;

using the data matching subunit to match the equipment operation state data with the equipment information according to the aggregation characteristics, and outputting corresponding matched storage equipment information of the equipment operation state data in the target storage frame information;

the step of using the feature aggregation subunit to aggregate the device running state feature and the storage device feature corresponding to each piece of storage device information to form a corresponding aggregate feature includes:

Utilizing the characteristic aggregation subunit to aggregate the first modality linear mapping characteristic and the second modality linear mapping characteristic to form a corresponding aggregation characteristic;

Determining a second modality linear mapping parameter for performing a second modality linear mapping; and performing second-modality linear mapping on the device running state characteristics and the storage device characteristics of each piece of storage device information according to the second-modality linear mapping parameters by using the characteristic aggregation subunit to form corresponding second-modality linear mapping characteristics; and obtaining a corresponding aggregation feature based on the second modality linear mapping feature;

The step of performing first-modality linear mapping on the device running state feature and the storage device feature of each piece of storage device information by using the feature aggregation subunit according to the first-modality linear mapping parameter to form a corresponding first-modality linear mapping feature includes:

weighting the first aggregation feature based on a first weight parameter included in the first modal linear mapping parameter to obtain a corresponding first weighted aggregation feature, and superposing the first weighted aggregation feature based on a first bias parameter included in the first modal linear mapping parameter to obtain a corresponding first modal linear mapping feature;

the step of performing second-mode linear mapping on the device running state feature and the storage device feature of each piece of storage device information according to the second-mode linear mapping parameter by using the feature aggregation subunit to form a corresponding second-mode linear mapping feature includes:

The step of acquiring the equipment operation state data of the oil gas production equipment in the target oil gas production system, performing feature mining on the equipment operation state data by utilizing a feature mining unit included in a monitoring data matching network, and outputting equipment operation state features corresponding to the equipment operation state data comprises the following steps:

aggregating the output states to represent the data segment characteristics to form the equipment operation state characteristics corresponding to the equipment operation state data;

the step of using the first data matching unit included in the monitoring data matching network to match subsystem information with the equipment operation state data according to the equipment operation state characteristics and outputting the matched storage subsystem information corresponding to the equipment operation state data in the target storage frame information includes:

2. The data stream arranging method applied to an oil and gas production system according to claim 1, wherein before the step of matching subsystem information of the equipment operation state data according to the equipment operation state characteristics by the first data matching unit included in the monitoring data matching network and outputting corresponding matched storage subsystem information of the equipment operation state data in target storage frame information, the data stream arranging method applied to an oil and gas production system further includes:

3. The method of arranging data streams applied to an oil and gas production system according to claim 2, wherein the step of aggregating the characteristics of the representative stored data segments corresponding to the representative stored data segments to form the characteristics of the storage devices corresponding to the storage device information comprises the steps of:

4. The data stream arranging method applied to an oil and gas production system according to any one of claims 1 to 3, wherein before the step of obtaining the equipment operation state data of the oil and gas production equipment in the target oil and gas production system and using the feature mining unit included in the monitoring data matching network to perform feature mining on the equipment operation state data and output the equipment operation state feature corresponding to the equipment operation state data, the data stream arranging method applied to the oil and gas production system further includes:

5. A data stream orchestration system comprising a processor and a memory, the memory being configured to store a computer program, the processor being configured to execute the computer program to implement the data stream orchestration method according to any one of claims 1-4, applied to an oil and gas production system.