Universal online predictive maintenance system for oil and gas pipelines

Abstract

The invention belongs to equipment diagnosis and maintenance systems, particularly relates to an online predictive maintenance system universal for all equipment of oil and gas pipelines, and aims to solve the problems of difficulty in anomaly positioning and low processing efficiency of a passive anomaly response mode of an existing fault diagnosis system. The method comprises the following steps: through corresponding sensors arranged on different devices or parts, according to a unified data acquisition interface and protocol, obtaining operation data and self-diagnosis data, and then selecting database types to carry out distributed storage and carry out unified device management; constructing an early warning model corresponding to each type of equipment or part based on the historical operation data, and setting early warning conditions for the operation data and the self-diagnosis data or each type or combination of the operation data and the self-diagnosis data; and based on the operation data and the self-diagnosis data, analyzing the operation trend of the equipment through an early warning model, and when the operation trend of the equipment triggers an early warning condition, sending out an early warning signal. According to the invention, equipment maintenance is converted into predictive maintenance management, and stable operation of the whole pipeline system is ensured.

Description

Background Art

[0002] At present, the oil and gas pipeline industry uses a monitoring and data acquisition system with industrial control computers as the core, namely the SCADA system, to achieve centralized monitoring, optimized operation and unified dispatch management of the entire pipeline. It can realize interlocking control logic, implement automatic distribution control, realize automatic start and stop and automatic protection sequence logic control of the entire station and the entire line, realize comprehensive remote control requirements, and achieve the goal of "unmanned operation, unmanned operation, and manned care". However, the station control system composed of PLC controls and manages the station field, and collects, detects and controls the main process parameters. The main processing data is process data, and the ability to collect and process equipment diagnostic data is low.

[0003] The current online remote diagnosis system for oil and gas pipeline industry equipment is composed of diagnostic system software provided by multiple equipment manufacturers or third-party manufacturers, mainly including: PLC / RTU diagnostic system, network equipment management system, metering remote diagnosis system, intelligent instrument management system, compressor unit / pump unit remote diagnosis system, etc.

[0004] Moreover, the current system does not have a unified data uploading standard, and the data uploaded by different manufacturers are inconsistent. This means that only a single device can be diagnosed, and it is impossible to form a systematic diagnosis for all devices.

[0005] At the same time, traditional fault diagnosis systems will only detect problems when components have problems, such as when a parameter exceeds the warning value. Maintenance personnel will then confirm whether it is caused by a process operation error, or check for causes such as on-site instrument failure, line short circuit, open circuit, PLC system software and hardware failure, etc.

[0006] This passive abnormal response mode has defects such as difficulty in locating abnormalities and low processing efficiency, and can no longer meet the current needs of the oil and gas pipeline industry for rapid abnormal location and processing. Summary of the invention

[0007] In order to solve the above-mentioned problems in the prior art, namely, the passive abnormal response mode of the existing fault diagnosis system has the problems of difficulty in abnormal location and low processing efficiency, the present invention provides an online predictive maintenance system universal for all equipment of oil and gas pipelines, the system comprising:

[0008] A data acquisition module is configured to acquire operation data and self-diagnosis data through corresponding sensors provided on different devices or components according to a unified data acquisition interface and protocol;

[0009] A data storage module, configured to select a corresponding database type for distributed storage and unified device management according to the operating data and self-diagnosis data;

[0010] a data analysis module configured to construct an early warning model corresponding to each device or component based on historical operation data, and to set early warning conditions for the operation data and self-diagnosis data or each one or more combinations thereof;

[0011] The early warning module is configured to analyze the equipment operation trend through the early warning model based on the operation data and the self-diagnosis data, and send out an early warning signal when the equipment operation trend triggers the early warning condition.

[0012] In some preferred embodiments, the system further comprises:

[0013] The maintenance report generation module is configured to issue a predictive maintenance report for a single device or all devices according to a set period through big data analysis algorithms.

[0014] In some preferred implementations, the warning condition is set as follows:

[0015] Based on factory specifications of the equipment or components used;

[0016] The corresponding early warning model is trained by historical operation data to obtain the historical operation trend of the equipment or parts; the historical operation trend includes the historical operation data and historical self-diagnosis data of the set time length before the equipment or parts fail;

[0017] When the operating data and the self-diagnosis data show a curve in which the similarity between the historical operating data and the historical self-diagnosis data of a set time length before the fault in the historical operating data is greater than the set fault threshold, the fault threshold is set as the warning condition.

[0018] In some preferred embodiments, the early warning module further includes:

[0019] For a single device, the historical operating data of all related equipment and components are analyzed and displayed in multiple dimensions. Low-latency indexing is supported for real-time data. Statistical analysis of operating data and self-diagnosis data is performed within seconds after access. This includes data aggregation, statistics, roll-up, drill-down, slicing, etc., and is displayed in the form of pie charts, line charts, bar charts, and rose charts.

[0020] In some preferred embodiments, the predictive maintenance report includes: equipment health assessment, potential fault identification, fault anomaly detection, comparative analysis, anomaly diagnosis and equipment threshold assessment.

[0021] In some preferred embodiments, the system further comprises a diagnostic management module, and a menu of a subsystem of the diagnostic management module is accessed by accessing the connection;

[0022] The subsystems of the diagnostic management module include:

[0023] PLC / RTU remote diagnosis management subsystem;

[0024] Network equipment remote management subsystem;

[0025] Metering system remote diagnosis and management subsystem;

[0026] Intelligent instrument remote diagnosis and management subsystem;

[0027] Unified security management platform.

[0028] In some preferred embodiments, the unified data collection interface and protocol supports the following data sources:

[0029] Smart devices, RTOS programmable devices and FPGA programmable devices;

[0030] Equipment control systems, PLC and DCS;

[0031] Acquisition monitoring system SCADA;

[0032] Production process execution management system MES;

[0033] Relational databases, Oracle and MySQL;

[0034] Data files, Json and CSV;

[0035] Distributed column storage databases, Hbase and MongoDB;

[0036] Supported transport protocols include: HTTPS, Web Socket, and MQTT.

[0037] In some preferred implementations, the system supports enterprise proprietary protocol parsing framework and data protocol configuration functions.

[0038] In some preferred implementations, the distributed storage specifically includes:

[0039] The distributed column database, relational database, distributed file subsystem, distributed real-time database and distributed time series database;

[0040] The distributed column database is used to store structured and semi-structured massive equipment operation data;

[0041] The relational database is used to store business data such as basic equipment information data, prediction and analysis result data, early warning and alarm data, and system metadata;

[0042] The distributed file subsystem is used to store statistical analysis index files;

[0043] The distributed real-time database is used for real-time monitoring, real-time analysis, and improving system operation efficiency services;

[0044] The distributed time series database is used to quickly write and read time series data, and is suitable for application scenarios where massive historical data is quickly displayed.

[0045] In some preferred embodiments, the data analysis module and the early warning module are implemented by distributed computing, including a distributed stream computing architecture, a distributed memory computing architecture, a computer learning algorithm library, and a distributed real-time analysis architecture;

[0046] The distributed stream computing architecture is used to realize real-time data monitoring, real-time alarm and early warning, and real-time computing;

[0047] Distributed memory computing architecture and machine learning algorithm library enable offline processing and calculation of massive data, and use statistical analysis algorithms and machine learning algorithms to build big data early warning models;

[0048] Distributed real-time analysis architecture implements index construction based on Lambda architecture, builds indexes in real time and offline modes, and realizes real-time statistical analysis of massive data.

[0049] Beneficial effects of the present invention:

[0050] (1) The present invention is to build an oil and gas pipeline into a safe, stable, efficient and highly automated pipeline, which can remotely control and manage the process systems and auxiliary systems of each process station and monitoring valve room at each dispatching and command center, and can also monitor the operating status and health status of pipeline-related equipment, so that equipment maintenance can be transformed into predictive maintenance management, thereby ensuring the stable operation of the entire pipeline system.

[0051] (2) This full-equipment diagnostic management system uses a diagnostic system based on the industrial Internet platform. The system integrates and improves the functions of existing decentralized diagnostic management systems. The system can achieve equipment monitoring, multi-dimensional analysis, status evaluation, fault diagnosis, intelligent early warning and other functions through big data and artificial intelligence models.

[0052] (3) The platform is based on a unified data interface and data access standard, which is conducive to the expansion of the platform in the future. The equipment diagnosis management system is positioned to unify the management of all equipment in the finished oil pipeline, collect and monitor the operating status and indicator data, analyze equipment health, diagnose faults and make maintenance decisions, so as to achieve remote predictive maintenance of all equipment, reduce equipment operating costs, extend equipment service life, and improve the overall equipment operation and maintenance response and processing efficiency of the pipeline network.

[0053] (4) This system realizes analysis functions such as fault potential identification, fault correlation analysis and diagnosis, provides intelligent applications for the equipment maintenance work of the oil and gas pipeline station operation team, and improves the accuracy of predictive maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0055] Figure 1 It is a structural block diagram of the online predictive maintenance system universal for all equipment of the oil and gas pipeline of the present invention;

[0056] Figure 2 It is a structural block diagram of the data acquisition module in an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings.

[0058] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0059] In order to more clearly explain the online predictive maintenance system for all equipment of the oil and gas pipeline of the present invention, the following is combined with Figure 1 Each module in the embodiment of the present invention is described in detail.

[0060] The online predictive maintenance system for all equipment of the oil and gas pipeline of the first embodiment of the present invention includes a data acquisition module, a data storage module, a data analysis module and an early warning module. The functional modules are described in detail as follows:

[0061] A data acquisition module is configured to acquire operation data and self-diagnosis data through corresponding sensors provided on different devices or components according to a unified data acquisition interface and protocol;

[0062] In this embodiment, the data collection method is as follows: Figure 2As shown, it includes parallel collection of DCS / PLC / RTU modules, flow detection modules, intelligent instrument equipment modules, dynamic equipment modules, leakage detection modules, network equipment modules and network security modules and transmission to the resource pool, and then transmission from the resource pool to the intelligent service predictive maintenance platform, i.e., this system. Applicable equipment for detection also includes temperature equipment, pressure equipment, flow equipment, liquid level equipment, and actuators. This step requires the collection of various types of data based on the characteristics of different equipment.

[0063] In this embodiment, the unified data collection interface and protocol supports the following data sources:

[0064] Smart devices, RTOS programmable devices and FPGA programmable devices;

[0065] Equipment control systems, PLC and DCS;

[0066] Acquisition monitoring system SCADA;

[0067] Production process execution management system MES;

[0068] Relational databases, Oracle and MySQL;

[0069] Data files, Json and CSV;

[0070] Distributed column storage databases, Hbase and MongoDB;

[0071] Supported transport protocols include: HTTPS, Web Socket, and MQTT.

[0072] The system proposed in this embodiment connects to the data interface protocols of multiple industry equipment, integrates multiple network transmission protocols, and implements the enterprise private data protocol analysis framework to achieve real-time collection of massive multi-source heterogeneous data. At the same time, the product provides batch data upload to achieve connection with existing data sources such as databases, file systems, and distributed column storage within the enterprise.

[0073] In this embodiment, the system supports the enterprise private protocol parsing framework and data protocol configuration functions, provides real-time and batch data source access, and helps users quickly connect and adapt to different types and models of devices.

[0074] A data storage module, configured to select a corresponding database type for distributed storage and unified device management according to the operating data and self-diagnosis data;

[0075] In this embodiment, the distributed storage specifically includes:

[0076] The distributed column database, relational database, distributed file subsystem, distributed real-time database and distributed time series database;

[0077] The distributed column database is used to store structured and semi-structured massive equipment operation data;

[0078] The relational database is used to store business data such as basic equipment information data, prediction and analysis result data, early warning and alarm data, and system metadata;

[0079] The distributed file subsystem is used to store statistical analysis index files;

[0080] The distributed real-time database is used for real-time monitoring, real-time analysis, and improving system operation efficiency services;

[0081] The distributed time series database is used to quickly write and read time series data, and is suitable for application scenarios where massive historical data (such as historical trend charts) are quickly displayed.

[0082] a data analysis module configured to construct an early warning model corresponding to each device or component based on historical operation data, and to set early warning conditions for the operation data and self-diagnosis data or each one or more combinations thereof;

[0083] This embodiment adopts B / S architecture to build a system platform, collect historical data of various equipment on site and use machine learning algorithms to build an early warning model for all equipment and components.

[0084] This embodiment is also used to perform unified data and equipment asset management, and provides unified management of enterprise equipment assets based on geographic information systems, cards, lists, etc., including management of enterprise regions and organizational structures, equipment types, equipment models, equipment overviews and details, equipment KPI summaries, and equipment asset analysis, helping users to conveniently and uniformly perform equipment management and basic analysis on the client.

[0085] In this embodiment, the warning condition is set as follows:

[0086] Based on factory specifications of the equipment or components used;

[0087] The corresponding early warning model is trained by historical operation data to obtain the historical operation trend of the equipment or parts; the historical operation trend includes the historical operation data and historical self-diagnosis data of the set time length before the equipment or parts fail;

[0088] When the operating data and the self-diagnosis data show a curve in which the similarity between the historical operating data and the historical self-diagnosis data of a set time length before the fault in the historical operating data is greater than the set fault threshold, the fault threshold is set as the warning condition.

[0089] In this embodiment, flexible warning configuration is provided, and threshold alarm conditions are configured according to the factory specifications of the equipment or components. When the monitored data reaches the threshold, an alarm is triggered according to the rules. At the same time, intelligent warning services such as warnings based on operating trends and abnormal detection warnings based on operating data are provided. Machine learning algorithms and massive historical data are used to build warning models, analyze equipment operating trends, trigger warnings before equipment fails, or detect and capture abnormal signals that occur during equipment operation to implement warnings.

[0090] The early warning module is configured to analyze the equipment operation trend through the early warning model based on the operation data and the self-diagnosis data, and send out an early warning signal when the equipment operation trend triggers the early warning condition.

[0091] In this embodiment, the early warning module further includes:

[0092] For a single device, the historical operating data of all related equipment and components are analyzed and displayed in multiple dimensions. Low-latency indexing is supported for real-time data. Statistical analysis of operating data and self-diagnosis data is performed within seconds after access. This includes data aggregation, statistics, roll-up, drill-down, slicing, etc., and is displayed in the form of pie charts, line charts, bar charts, and rose charts.

[0093] In this embodiment, the data analysis module and the early warning module are implemented through distributed computing, including a distributed stream computing architecture, a distributed memory computing architecture, a computer learning algorithm library, and a distributed real-time analysis architecture;

[0094] The distributed stream computing architecture is used to realize real-time data monitoring, real-time alarm and early warning, and real-time computing;

[0095] Distributed memory computing architecture and machine learning algorithm library enable offline processing and calculation of massive data, and use statistical analysis algorithms and machine learning algorithms to build big data early warning models;

[0096] Distributed real-time analysis architecture implements index construction based on Lambda architecture, builds indexes in real time and offline modes, and realizes real-time statistical analysis of massive data.

[0097] This embodiment uses a distributed stream computing architecture to calculate the collected field equipment data and upload it in real time through a data gateway, calls a prediction service interface to process and analyze the equipment operation trend, detects the trend of fault occurrence, triggers an early warning before the equipment fails, or detects and captures abnormal signals that appear during the operation of the equipment to achieve early warning. This embodiment is also used to detect and capture abnormal signals that appear during the operation of the equipment to achieve early warning.

[0098] In this embodiment, a predictive analysis model for equipment and parts is constructed to efficiently and accurately detect the trend of fault occurrence, realize advanced analysis functions such as fault potential identification, fault correlation analysis and diagnosis, provide intelligent applications for the equipment maintenance work of the operation team, and provide a technical basis for the construction of a predictive maintenance system based on the equipment operation status. The implementation of predictive maintenance will greatly reduce the economic losses caused by failures, save costs, and improve equipment operation efficiency and energy efficiency. Predictive maintenance provides the following basic applications, and can provide business-specific model construction and application deployment for different industry businesses and application scenarios.

[0099] In this embodiment, the system further includes:

[0100] The maintenance report generation module is configured to issue a predictive maintenance report for a single device or all devices according to a set period through big data analysis algorithms.

[0101] In this embodiment, normal working data of equipment of the same type and working condition is obtained from the database, and the normal working data is fitted into a normal working curve; the operating data and self-diagnosis parameters of the current single equipment are compared with the normal working curve; if the curve similarity is lower than the set working threshold, it indicates that maintenance is required.

[0102] In this embodiment, the predictive maintenance report includes: equipment health assessment, potential fault identification, fault anomaly detection, comparative analysis, anomaly diagnosis and equipment threshold assessment.

[0103] In this embodiment, the system further comprises a diagnosis management module, and a menu of a subsystem of the diagnosis management module is entered by accessing the connection;

[0104] The subsystems of the diagnostic management module include:

[0105] PLC / RTU remote diagnosis management subsystem;

[0106] Network equipment remote management subsystem;

[0107] Metering system remote diagnosis and management subsystem;

[0108] Intelligent instrument remote diagnosis and management subsystem;

[0109] Unified security management platform.

[0110] This embodiment implements a comprehensive monitoring module through an early warning module, a maintenance report generation module, and a diagnostic management module. The comprehensive monitoring module helps users to fully understand the connected devices, monitor the key information of the devices such as offline status, operating status, and health, monitor the operating data of the device sensors in real time, and provide early warning alarms, historical data query and download, multi-dimensional statistical analysis and other functions.

[0111] The system proposed in this embodiment is based on mature big data technology and machine learning algorithms, and multiple algorithms are adjusted and optimized for the field of Internet of Things. Multiple predictive maintenance and optimization improvement models are integrated into the product. It has predictive analysis and optimization improvement model templates in multiple industry fields, which can quickly realize model construction and tuning.

[0112] In this embodiment, it also includes a user management and permission module, which is configured to provide the platform with user, role, and permission management functions. All business systems implement unified login, authentication, and authorization functions through the user center. It supports the user's permission configuration function, and realizes the isolation of user display permissions and data permissions through configuration; users can only enter the business system allowed by permissions, and users can only view and manage the corresponding devices within the scope of permissions.

[0113] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process and related instructions of the storage device and processing device described above can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0114] Those skilled in the art should be able to appreciate that the modules and method steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, computer software or a combination of the two, and the programs corresponding to the software modules and method steps can be placed in random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium known in the technical field. In order to clearly illustrate the interchangeability of electronic hardware and software, the composition and steps of each example have been generally described in the above description according to the function. Whether these functions are performed in electronic hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0115] The terms "first", "second", etc. are used to distinguish similar objects rather than to describe or indicate a particular order or sequence.

[0116] The term "comprise" or any other similar term is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus / device that includes a list of elements includes not only those elements but also other elements not expressly listed, or also includes elements inherent to such process, method, article, or apparatus / device.

[0117] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. An online predictive maintenance system for all equipment in oil and gas pipelines, characterized in that: The system comprises: A data acquisition module is configured to acquire operation data and self-diagnosis data through corresponding sensors provided on different devices or components according to a unified data acquisition interface and protocol; A data storage module, configured to select a corresponding database type for distributed storage and unified device management according to the operating data and self-diagnosis data; a data analysis module configured to construct an early warning model corresponding to each device or component based on historical operation data, and to set early warning conditions for the operation data and self-diagnosis data or each one or more combinations thereof; The early warning module is configured to analyze the equipment operation trend through the early warning model based on the operation data and the self-diagnosis data, and send out an early warning signal when the equipment operation trend triggers the early warning condition.

2. The online predictive maintenance system for all equipment in the oil and gas pipeline according to claim 1 is characterized in that: The system further comprises: The maintenance report generation module is configured to issue a predictive maintenance report for a single device or all devices according to a set period through big data analysis algorithms.

3. The online predictive maintenance system for all equipment in the oil and gas pipeline according to claim 2 is characterized in that: The method for setting the warning condition is as follows: Based on factory specifications of the equipment or components used; The corresponding early warning model is trained by historical operation data to obtain the historical operation trend of the equipment or parts; the historical operation trend includes the historical operation data and historical self-diagnosis data of the set time length before the equipment or parts fail; When the operating data and the self-diagnosis data show a curve in which the similarity between the historical operating data and the historical self-diagnosis data of a set time length before the fault in the historical operating data is greater than the set fault threshold, the fault threshold is set as the warning condition.

4. The online predictive maintenance system for all equipment in the oil and gas pipeline according to claim 3 is characterized in that: The early warning module further includes: For a single device, the historical operating data of all related equipment and components are analyzed and displayed in multiple dimensions. Low-latency indexing is supported for real-time data. Statistical analysis of operating data and self-diagnosis data is performed within seconds after access. This includes data aggregation, statistics, roll-up, drill-down, slicing, etc., and is displayed in the form of pie charts, line charts, bar charts, and rose charts.

5. The online predictive maintenance system for all equipment in the oil and gas pipeline according to claim 4 is characterized in that: The predictive maintenance report includes: equipment health assessment, potential fault identification, fault anomaly detection, comparative analysis, anomaly diagnosis and equipment threshold assessment.

6. The online predictive maintenance system for all equipment in the oil and gas pipeline according to claim 5 is characterized in that: The system also includes a diagnostic management module, and a menu of a subsystem of the diagnostic management module is accessed by accessing the connection; The subsystems of the diagnostic management module include: PLC / RTU remote diagnosis management subsystem; Network equipment remote management subsystem; Metering system remote diagnosis and management subsystem; Intelligent instrument remote diagnosis and management subsystem; Unified security management platform.

7. The online predictive maintenance system for all equipment in the oil and gas pipeline according to claim 1 is characterized in that: The unified data collection interface and protocol supports the following data sources: Smart devices, RTOS programmable devices and FPGA programmable devices; Equipment control systems, PLC and DCS; Acquisition monitoring system SCADA; Production process execution management system MES; Relational databases, Oracle and MySQL; Data files, Json and CSV; Distributed column storage databases, Hbase and MongoDB; Supported transport protocols include: HTTPS, Web Socket, and MQTT.

8. The online predictive maintenance system for all equipment in the oil and gas pipeline according to claim 1 is characterized in that: The system supports the enterprise private protocol parsing framework and data protocol configuration functions.

9. The online predictive maintenance system for all equipment in the oil and gas pipeline according to claim 1 is characterized in that: The distributed storage specifically includes: The distributed column database, relational database, distributed file subsystem, distributed real-time database and distributed time series database; The distributed column database is used to store structured and semi-structured massive equipment operation data; The relational database is used to store business data such as basic equipment information data, prediction and analysis result data, early warning and alarm data, and system metadata; The distributed file subsystem is used to store statistical analysis index files; The distributed real-time database is used for real-time monitoring, real-time analysis, and improving system operation efficiency services; The distributed time series database is used to quickly write and read time series data, and is suitable for application scenarios where massive historical data is quickly displayed.

10. The online predictive maintenance system for all equipment in the oil and gas pipeline according to claim 1 is characterized in that: The data analysis module and the early warning module are implemented through distributed computing, including a distributed stream computing architecture, a distributed memory computing architecture, a computer learning algorithm library, and a distributed real-time analysis architecture; The distributed stream computing architecture is used to realize real-time data monitoring, real-time alarm and early warning, and real-time computing; Distributed memory computing architecture and machine learning algorithm library enable offline processing and calculation of massive data, and use statistical analysis algorithms and machine learning algorithms to build big data early warning models; Distributed real-time analysis architecture implements index construction based on Lambda architecture, builds indexes in real time and offline modes, and realizes real-time statistical analysis of massive data.

Images