Cloud edge collaborative architecture system and equipment of gas turbine operation and maintenance platform and storage medium

Abstract

The invention discloses a cloud-side collaborative architecture system of a gas turbine operation and maintenance platform, which is composed of a cloud system platform and an edge system platform and is used for providing the following cloud-side collaborative services: data collaboration: after real-time data acquisition and processing are performed based on the edge system platform, the cloud-side collaborative architecture system is used for carrying out cloud-side collaborative processing; a part of data is selected to support real-time index updating of a foreground interface of the power plant and the cloud system, and other unselected parts are used for supporting an intelligent application model of real-time calculation; business application management collaboration: selecting a specified part of application data to carry out cloud business application arrangement, issuing the other part of application data to an edge system, and arranging applications through business application middleware; and service collaboration: providing calling interaction of the cloud capability for the edge application based on the cloud system platform, sinking the cloud capability to the edge system platform, and locally providing services for the edge application. According to the invention, the data processing efficiency can be obviously improved, the network bandwidth pressure is reduced, the resource configuration is optimized, and the flexibility and security of the system are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of operation and maintenance platform system architecture, and specifically to a cloud-edge collaborative architecture system, equipment and storage medium for a gas turbine operation and maintenance platform. Background Art

[0002] With the development of industrial Internet and cloud computing technologies, modern power plants, especially gas turbine power plants, are increasingly relying on intelligent operation and maintenance methods to improve operational efficiency, reduce maintenance costs and ensure production safety. The common practice of existing intelligent operation and maintenance is to build an intelligent operation and maintenance platform to intelligently realize fault warning and diagnosis management of gas turbines.

[0003] In the prior art, there is a patent with publication number CN118446680A: a gas turbine power plant intelligent operation and maintenance device system and operation method, which establishes a high-precision performance analysis model for the gas turbine, an intelligent early warning model for key components of the gas turbine, and an analysis data model, thereby realizing early warning of abnormal conditions of the gas turbine and solving the problem of high fault warning omission or false alarm rate in the existing operation and maintenance platform.

[0004] However, existing intelligent operation and maintenance system platforms, including the aforementioned patents, still have the following problems in actual use:

[0005] (1) Existing intelligent operation and maintenance systems usually centralize all data in the cloud for processing and analysis, resulting in high data transmission and processing delays, which cannot meet the needs of real-time monitoring and rapid response;

[0006] (2) Large amounts of raw data are transmitted from edge devices to the cloud, which puts enormous pressure on network bandwidth. Especially under poor network conditions, data loss or transmission interruption is likely to occur;

[0007] (3) Low resource utilization: Existing systems often rely on a single cloud computing resource and fail to fully utilize the computing power of edge devices, resulting in low resource utilization and increased operation and maintenance costs.

[0008] (4) Inflexible application deployment: Existing intelligent operation and maintenance systems usually adopt a centralized architecture, which makes it difficult to flexibly adjust the type and number of edge applications according to actual needs, limiting the adaptability and scalability of the system.

[0009] (5) Data security issues: All data are stored in the cloud. Once a data leak or attack occurs, it may lead to serious security issues.

[0010] To this end, this application specifically proposes a cloud-edge collaborative architecture system for a gas turbine operation and maintenance platform to solve the above technical problems. Summary of the Invention

[0011] The main purpose of the present invention is to provide a cloud-edge collaborative architecture system for a gas turbine operation and maintenance platform to solve the technical problems raised in the background technology.

[0012] The present invention adopts the following technical solutions to solve the above technical problems:

[0013] A cloud-edge collaborative architecture system for a gas turbine operation and maintenance platform, consisting of a cloud system platform and an edge system platform, is used to provide cloud-edge collaborative services including data collaboration, business application management collaboration, and service collaboration to perform industrial data access, protocol parsing and data preprocessing, and edge analysis application operations.

[0014] Data collaboration: After real-time data collection and processing based on the edge system platform, some data is selected to support real-time indicator updates on the front-end interfaces of the power plant and cloud system. The remaining unselected data is used to support intelligent application models for real-time calculations. The model calculation results are sent to the front-end interfaces of the power plant and cloud system. The edge system platform regularly transmits incremental power plant production data to the cloud system platform, which processes and analyzes the data transmitted by the edge system.

[0015] Business application management collaboration selects a specified portion of application data for cloud-based business application orchestration, while sending the remaining portion of application data to the edge system. Through business application middleware and application orchestration, both real-time and periodic computing applications can run efficiently.

[0016] Service collaboration is based on the cloud system platform to provide edge applications with cloud capability call interaction, and sink cloud capabilities to the edge system platform to provide services for edge applications locally.

[0017] Preferably, the specific management steps of the business application management collaboration include: allowing the data to be processed and analyzed by the cloud system platform, and then developing, training, and reasoning intelligent applications, selecting some data applications for deployment in the cloud, and performing application orchestration according to different application requirements, and sending other unselected data applications to the edge system platform. Through the business application middleware and the orchestration of applications, real-time computing applications and periodic computing applications can run efficiently, which is used for unified management of cloud applications and edge systems.

[0018] Preferably, the cloud system platform consists of a resource service layer IaaS, a platform layer PaaS and an application layer SaaS, wherein:

[0019] The resource service layer, IaaS, provides computing, storage, networking, and operation and maintenance services based on basic resources to meet the business needs and different deployment methods of the PaaS and SaaS layers.

[0020] Platform layer PaaS, including general and specialized capabilities, is used to specify service functions to meet the application development needs of the SaaS layer and power plant-side model development and training;

[0021] Application layer SaaS is used to provide new industrial applications based on business needs and build designated business applications.

[0022] Preferably, the computing, storage, network and operation and maintenance services of the resource service layer IaaS specifically include:

[0023] Computing services: Provide multi-engine computing services including cloud hosts and containers, support application systems of different architecture types, and manage them through a unified service interface to ensure a consistent user experience;

[0024] Storage service: Provides production-level cloud hard disk services and ensures data security through snapshot, full and incremental backup mechanisms;

[0025] Network services: Based on software-defined networking technology, we provide virtualized network services and VPC private network services to ensure strict network isolation between different projects. We also provide network services including virtual routers, firewalls, public IP addresses, and security groups. We also support L4-L7 load balancing services to meet users' network connectivity needs in different application scenarios.

[0026] Preferably, the overall technical architecture of the platform layer PaaS includes the following system layers:

[0027] The foundational layer uses mainstream container orchestration and isolation technologies to isolate applications from each other and from the host operating system, ensuring that backend function upgrades are sustainable.

[0028] The engine adaptation layer adapts to diversified storage and computing engines to maximize resource utilization;

[0029] The platform layer uses ElasticSearch to achieve efficient full-text search, provides efficient caching services through Redis / Mongo, improves data loading efficiency, and builds a real-time message channel through Kafka to achieve real-time data access.

[0030] The service layer divides internal gateways and external gateways according to business forms, provides service authentication methods according to different business scenarios, and uses service clusters to provide callers with stable and reliable service entry in high-concurrency scenarios;

[0031] The presentation layer is encapsulated and extended based on vue and ElementUi.

[0032] Preferably, the application layer SaaS adopts the SOA framework, takes advantage of loose coupling between layers and high cohesion within layers, and realizes business object componentization and unified service calls through business abstraction and mapping, specifically including:

[0033] The component layer is used for component-based abstraction and lowest-granularity assembly of system business objects;

[0034] The service layer integrates and encapsulates relevant business components in a highly granular manner, providing a standard calling interface for designated business applications for direct invocation by business applications.

[0035] The application layer includes specific and achievable business applications required by the system business and management levels;

[0036] The system business starts at the service layer, and uses the transaction manager of the database to perform transaction control processing on the relevant services of the service layer.

[0037] Preferably, the edge system platform adopts an edge computing architecture, which rapidly collects, stores and processes data through edge nodes deployed at the edge of the network to reduce the pressure on network bandwidth;

[0038] A time series database is deployed within the edge system platform to collect and store unit data, and upload the unit data to the cloud as needed to provide a sample set for training complex algorithm models on the platform side. The trained edge algorithm service is stored in a lightweight mode in the local power plant to meet the application needs of the power plant side.

[0039] The edge algorithm service is used to open up data interaction between the edge system and the cloud system. Through real-time API services, the final calculation results of the edge system are stored in the local database and pushed to the center side at regular intervals.

[0040] The edge system platform sets customized task scheduling for massive data in an offline manner to complete the regular synchronization of massive data from the edge system to the center side;

[0041] The edge system platform is internally provided with an algorithm service center, which defines different versions of services for the same model, and is used to obtain API services published on the center side in real time, and to call them in real time to obtain the data required for participating in edge system calculations.

[0042] Preferably, the edge system platform includes the following system supports:

[0043] The management-side application system is deployed in the production management area of ​​the power plant to form cloud-edge collaboration with the cloud system, thereby providing the power plant with intelligent applications that assist in production and management.

[0044] Production-side application system support, as a closed-loop feedback function module of the edge system platform, is deployed in the combined cycle control system DCS of the gas turbine, or deployed in the TCS in combination with the gas turbine control system. It is used to arrange performance optimization control servers in parallel with the engineer station based on the TCS / DCS Ethernet, and deploy reverse isolation devices between the production control area and the management information area.

[0045] Preferably, the edge algorithm service training step of the edge system platform includes:

[0046] S1. Perform data preprocessing operations, including noise filtering and data cleaning, on the update data collected by the edge nodes according to preset requirements;

[0047] S2. Based on the same preset requirements, query the data records of the generator sets stored in the time series database over time, and combine the query data with the preprocessed data set to form a unified data set Q;

[0048] S3. For dataset Q, determine the amount of data in the dataset. If the amount is greater than a preset threshold, consider it a large dataset and use batch processing techniques and a greedy algorithm to optimize the resource data to ensure that the amount of data in the dataset is within the specified threshold.

[0049] S4. Deploy the local environment, build an initial data decision tree model based on the dataset, traverse the dataset Q, and train the algorithm model using the decision tree classification and regression algorithm. During the training process, control the complexity of the decision tree by setting the max_depth and min_samples_leaf parameters to prevent overfitting.

[0050] On the other hand, the present invention also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor constructs a cloud-edge collaborative architecture system of the gas turbine operation and maintenance platform as described above.

[0051] On the other hand, the present invention further discloses a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor constructs a cloud-edge collaborative architecture system of the gas turbine operation and maintenance platform as described above.

[0052] As can be seen from the above technical solutions, the present invention provides a cloud-edge collaborative architecture system for a gas turbine operation and maintenance platform. Compared with the existing technology, the present invention has the following advantages:

[0053] 1. By establishing a data collaboration mechanism between the cloud system platform and the edge system platform, the present invention can update the front-end interface indicators of the power plant and the cloud system in real time, thereby improving data processing efficiency and response speed, and enhancing the operating performance of the overall system.

[0054] 2. By setting up a business application management collaboration mechanism, the present invention can achieve effective integration of cloud and edge computing resources and flexible deployment of applications, thereby optimizing resource allocation, accelerating business processes, and enhancing the adaptability and scalability of the system.

[0055] 3. The present invention sets up a service collaboration mechanism so that cloud capabilities can be transferred to the edge, providing localized service support for edge applications. This not only reduces dependence on the central network, but also improves the service quality of edge applications.

[0056] 4. The present invention can significantly improve data processing efficiency, reduce network bandwidth pressure, optimize resource allocation, and enhance system flexibility and security by setting up cloud-edge collaborative services including data collaboration, business application management collaboration, and service collaboration.

[0057] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become easy to understand through the following description. Of course, it is not necessary to achieve all of the above-mentioned advantages simultaneously in order to implement any product of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are intended to explain the present invention and do not constitute an undue limitation of the present invention. In the accompanying drawings:

[0059] Figure 1 This is a schematic diagram of the cloud-edge collaborative architecture of the gas turbine intelligent operation and maintenance cloud platform of the present invention;

[0060] Figure 2 This is a schematic diagram of the cloud system architecture of the gas turbine intelligent operation and maintenance cloud platform of the present invention;

[0061] Figure 3 This is a schematic diagram of the overall technical architecture of the PaaS layer of the present invention;

[0062] Figure 4 This is a schematic diagram of the functional architecture of the application layer of the gas turbine intelligent operation and maintenance cloud platform of the present invention;

[0063] Figure 5 This is a schematic diagram of the technical architecture of the application layer of the gas turbine intelligent operation and maintenance cloud platform of the present invention;

[0064] Figure 6 This is a logic diagram for realizing the platform B / S structure of the present invention;

[0065] Figure 7 A schematic diagram of the platform development and design architecture of the present invention;

[0066] Figure 8This is a schematic diagram of the edge system architecture of the gas turbine intelligent operation and maintenance cloud platform of the present invention. DETAILED DESCRIPTION

[0067] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. In the absence of conflict, the embodiments in this application and the features in the embodiments can be combined with each other. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0068] In the embodiment, see Figures 1 to 8 .

[0069] like Figure 1 As shown, a cloud-edge collaborative architecture system for a gas turbine operation and maintenance platform proposed in an embodiment of the present invention is composed of a cloud system platform and an edge system platform. It is used to provide cloud-edge collaborative services including data collaboration, business application management collaboration, and service collaboration to perform industrial data access, protocol parsing and data preprocessing, and edge analysis application operations, wherein:

[0070] (1) Data collaboration: After real-time data collection and processing based on the edge system platform, some data are selected to support the real-time indicator update of the front-end interface of the power plant and the cloud system. The remaining unselected data are used to support the intelligent application model of real-time calculation. The model calculation results are sent to the front-end interface of the power plant and the cloud system. The edge system platform regularly transmits the incremental production data of the power plant to the cloud system platform, and the cloud system platform is used to process and analyze the data transmitted by the edge system;

[0071] (2) Business application management collaboration: select a specified part of the application data for cloud business application orchestration, and send the other part of the application data to the edge system. Through the business application middleware and the orchestration of the application, the real-time computing application and the periodic computing application can run efficiently. The specific management steps include: after the data is processed and analyzed by the cloud system platform, the development, training, and reasoning of intelligent applications are carried out. Select part of the data application to be deployed in the cloud, and perform application orchestration according to different application requirements. The other unselected part of the data application is sent to the edge system platform. Through the business application middleware and the orchestration of the application, the real-time computing application and the periodic computing application can run efficiently. It is used to uniformly manage cloud applications and edge systems. At the same time, the application type and number of the edge system can be changed according to the actual needs of different power plants.

[0072] It should be noted that the business application management collaboration mechanism can achieve effective integration of cloud and edge computing resources and flexible deployment of applications, thereby optimizing resource allocation, accelerating business processes, and ultimately enhancing the adaptability and scalability of the system.

[0073] (3) Service collaboration: The cloud system platform is used to provide edge applications with cloud capabilities for call interaction, and the cloud capabilities are transferred to the edge system platform to provide services for edge applications locally;

[0074] It should be noted that the setting of the service collaboration mechanism here enables cloud capabilities to be transferred to the edge, providing localized service support for edge applications. This not only reduces dependence on the central network, but also improves the service quality of edge applications, ensuring a good user experience even in poor network conditions.

[0075] In summary, through the above mechanism, data processing efficiency can be significantly improved, network bandwidth pressure can be reduced, resource allocation can be optimized, and system flexibility and security can be enhanced, thereby better meeting the needs of intelligent operation and maintenance of modern power plants. At the same time, by establishing a data collaboration mechanism between the cloud system platform and the edge system platform, it can also play a role in updating the front-end interface indicators of the power plant and cloud system in real time, thereby achieving the effect of improving data processing efficiency and response speed, and finally improving the operating efficiency of the overall system.

[0076] Further, such as Figure 2 As shown in the figure, the cloud system platform consists of the resource service layer IaaS, the platform layer PaaS and the application layer SaaS:

[0077] (1) Resource service layer IaaS, which provides computing, storage, network, and operation and maintenance services based on basic resources to meet the business needs and different deployment methods of PaaS and SaaS layers;

[0078] The computing, storage, network, and operation and maintenance services of the resource service layer IaaS specifically include:

[0079] Computing services: Provide multi-engine computing services including cloud hosts and containers, support application systems of different architecture types, and manage them through a unified service interface to ensure a consistent user experience;

[0080] Storage service: Provides production-level cloud hard disk services and ensures data security through snapshot, full and incremental backup mechanisms;

[0081] Network Services: Based on software-defined networking technology, we provide virtualized network services, including VPC private network services, to ensure strict network isolation between different projects. We also provide network services including virtual routers, firewalls, public IP addresses, and security groups. We also support L4-L7 load balancing services to meet users' network connectivity needs in different application scenarios.

[0082] It should be noted that the diverse computing, storage, network, and operation and maintenance services provided by the IaaS layer can meet the application requirements of different architecture types and ensure data security and network connection stability, thereby achieving the goal of building an efficient and reliable cloud environment and supporting a wider range of industrial application scenarios.

[0083] (2) Platform layer PaaS, including general and professional capabilities, is used to specify service functions to meet the application development needs of the SaaS layer and the development and training of power plant side models, such as Figure 3 As shown in the figure, the overall technical architecture of the platform layer PaaS includes the following system layers:

[0084] The foundational layer uses mainstream container orchestration and isolation technologies to isolate applications from each other and from the host operating system, ensuring that backend functionality upgrades are sustainable.

[0085] The engine adaptation layer adapts to diversified storage and computing engines to maximize resource utilization;

[0086] The platform layer uses ElasticSearch to achieve efficient full-text search, provides efficient caching services through Redis / Mongo, improves data loading efficiency, and builds a real-time message channel through Kafka to achieve real-time data access.

[0087] The service layer divides internal gateways and external gateways according to business forms, provides service authentication methods according to different business scenarios, and uses service clusters to provide callers with stable and reliable service entry in high-concurrency scenarios;

[0088] Presentation layer, encapsulated and extended based on vue and ElementUi;

[0089] It should be noted that the design of the PaaS layer technical architecture, especially the application of container orchestration and isolation technology in the base layer, and the engine adaptation layer's support for multi-source storage and computing engines, effectively enhance the platform's flexibility and performance. The service layer, based on technologies such as ElasticSearch, further enhances the speed and accuracy of information retrieval, facilitating rapid response in complex business scenarios.

[0090] (3) Application layer SaaS, used to provide new industrial applications based on business needs and build designated business applications, including reference Figure 4 and Figure 5 To ensure the platform's horizontal and vertical expansion capabilities, the application layer SaaS adopts the SOA framework. It leverages loose coupling between layers and high cohesion within layers. Through business abstraction and mapping, it implements business object componentization and unified service calls. This fully considers the system's scalability, reusability, and configurability, reducing development and maintenance costs and enabling the system to adapt to changing needs, quickly and flexibly meeting the needs of business changes. Specifically, it includes:

[0091] The component layer is used for component-based abstraction and lowest-granularity assembly of system business objects;

[0092] The service layer integrates and encapsulates relevant business components in a highly granular manner, providing a standard calling interface for designated business applications for direct invocation by business applications.

[0093] The application layer includes specific and achievable business applications required by the system business and management levels;

[0094] System operations start at the service layer, and the database transaction manager is used to perform transaction control on related services at the service layer.

[0095] In addition, the platform application uses the Browser / Server structure (B / S structure for short). The B / S structure is one of the most mature and flexible architectures of current application systems. Users only need to use the Browser to access the system and perform business operations without installing special client tools to use the system. The specific implementation logic is as follows: Figure 6 As shown, the platform adopts the principle of loose coupling and layered design, which is mainly divided into UI (User Interface) layer, backend JVM, and data service layer. Figure 7 As shown, specifically including:

[0096] (a)UI layer

[0097] Refers to the user interface, which is the interface directly presented to users by modules such as early warning diagnosis, energy efficiency monitoring, and energy efficiency diagnosis. The UI layer is mainly implemented using the Vue framework, Element-UI, and Echarts components;

[0098] (b) Backend JVM

[0099] The backend JVM is the implementation of the business logic of the external business service functions provided by this platform, which is implemented using microservice components such as SpringBoot, SpringSecurity, and Jwt;

[0100] The SpringBoot microservice framework is used as the processing layer development technology for business logic because the application services required by each power plant are not completely consistent. Later, when deploying on the power plant side, service selection and combination packaging can be performed. On the other hand, the SpringBoot architecture can also be used for distributed deployment of services based on the service application situation. For example, if the early warning module has a large number of service calls, which will affect the overall system performance, the early warning module can be deployed separately without affecting the use of other business modules.

[0101] (c) Data Service Layer

[0102] The data service layer provides an interface for accessing database data and is directly called by the business processing layer or service layer. The data service layer is mainly implemented through Mybatis components and related components provided by the data center (such as real-time computing, DataX, etc.). Transaction processing begins at the service layer, using the database transaction manager to perform transaction control on related services of the service layer.

[0103] It should be noted that the SaaS layer utilizes the SOA framework to implement component-based and service-oriented business logic design. This layered abstraction approach helps simplify the integration workload during software development while ensuring the independent operation of each functional module, significantly reducing system maintenance costs. It also facilitates the rapid iteration and release of new features or improvements, enabling enterprises to more agilely respond to market changes and technological challenges.

[0104] (4) General data services: including data collection, data storage, data governance, data calculation and other capabilities; there are two main ways to collect data in this project. One is to directly obtain data from the database of the data platform when the power plant already has a data platform and has carried out data governance; the other is to obtain data from different business systems on site if the power plant does not have a data platform and perform corresponding data governance, such as obtaining production data from the PI database and obtaining unit maintenance and fault data from the maintenance system; the data governance system takes the metadata of the production and maintenance data of the gas turbine power plant as the core, and drives the standardization, normalization, visualization and integrated management of data. Management, to achieve intelligent extraction of data standards and integrated governance of global data; for multiple data sources, including structured, semi-structured, and unstructured data, to achieve global heterogeneous data integration and build a standard and normative data system; some indicators and intelligent models in this project require real-time calculation, that is, timely calculation of the collected key indicator data; the real-time data processing engine receives real-time data sent by the real-time data interface, and performs real-time data calculations on each measurement point according to pre-defined data rules or intelligent models, and promptly feeds back the calculation results to the front-end interface so that the operation monitoring personnel can understand the operating status of the power plant units and take timely and efficient response measures;

[0105] (5) Construction of industrial algorithm functions: The industrial algorithm function is one of the core business functions of the gas turbine intelligent operation and maintenance platform. The platform provides a large number of models and algorithm integrations, including: clustering, classification, decision tree, neural network, information entropy (IV), Pearson correlation coefficient, Spearman correlation coefficient, chi-square test, Gini index, Euclidean distance, cosine similarity, Pearson correlation coefficient, Hamming distance, Jaccard similarity coefficient, edit distance, KL divergence and other algorithms; and provides model development, service, management and other services to meet the long-term development and operation needs of gas turbine intelligent models;

[0106] (6) Developer Services: To meet the development or operation of the underlying and business applications of the gas turbine intelligent operation and maintenance platform, the platform consists of container services, microservice development frameworks, and application development functions;

[0107] (7) API services: In addition to industrial algorithm research and development, the platform also opens a variety of API service interfaces in other directions to meet the needs of users for other customized functions. By opening API services, the platform has scalability. Users can combine the data, events, and functions in the platform with their own applications through API service interfaces, making the platform more closely connected with users' daily work; such as sharing model data with edge systems;

[0108] Further, such as Figure 8 As shown in the figure, the edge system platform adopts edge computing architecture to provide functions such as access to massive industrial data, protocol parsing and data preprocessing, and edge analysis applications. It quickly collects, stores, and processes data through edge nodes deployed at the edge of the network to reduce the pressure on network bandwidth.

[0109] The edge system platform adopts different data access methods based on the power plant data collection; provides different edge data processing functions based on the power plant side function deployment method, adopts local deployment method to improve computing efficiency and reduce network bandwidth requirements; the edge system provides cloud-edge collaborative services; the security system provides security protection for each layer of the platform; the operation and maintenance system provides maintenance and scheduling for each layer of the platform

[0110] A time series database is deployed within the edge system platform to collect and store unit data. This data is then uploaded to the cloud as needed to provide sample sets for training complex algorithm models on the platform side. The trained edge algorithm service is stored in a lightweight mode at the local power plant to meet plant-side application requirements.

[0111] The edge algorithm service is used to facilitate data interaction between edge systems and cloud systems. Through real-time API services, the final calculation results of the edge system are stored in the local database and pushed to the center side at regular intervals.

[0112] The edge system platform sets customized task scheduling for massive data in an offline manner, completing the scheduled synchronization of massive data from the edge system to the center side;

[0113] An algorithm service center is set up inside the edge system platform, which defines different versions of services for the same model. It is used to obtain API services published on the center side in real time, and call them in real time to obtain the data required for participating in edge system calculations. At this time, the existence of the algorithm service center ensures the effectiveness of model version control, facilitates continuous optimization of algorithm performance, and also promotes data flow between edge and cloud, which is conducive to forming a closed-loop learning mechanism and continuously improving prediction accuracy and service level.

[0114] Among them, the edge system platform includes the following system supports:

[0115] The management-side application system is deployed in the production management area of ​​the power plant to form cloud-edge collaboration with the cloud system, thereby providing the power plant with intelligent applications that assist in production and management.

[0116] Production-side application system support, as a closed-loop feedback function module of the edge system platform, is deployed in the combined cycle control system DCS of the gas turbine, or deployed in the TCS in conjunction with the gas turbine control system. It is used to arrange performance optimization control servers in parallel with the engineer station based on the TCS / DCS Ethernet network, and to deploy reverse isolation devices between the production control area and the management information area.

[0117] At this time, the edge algorithm service training steps of the edge system platform include:

[0118] S1. Perform data preprocessing operations, including noise filtering and data cleaning, on the update data collected by the edge nodes according to preset requirements;

[0119] S2. Based on the same preset requirements, query the data records of the generator sets stored in the time series database over time, and combine the query data with the preprocessed data set to form a unified data set Q;

[0120] S3. For dataset Q, determine the amount of data in the dataset. If the amount is greater than a preset threshold, consider it a large dataset and use batch processing techniques and a greedy algorithm to optimize the resource data to ensure that the amount of data in the dataset is within the specified threshold.

[0121] S4. Deploy the local environment, build an initial data decision tree model based on the dataset, traverse the dataset Q, and train the algorithm model using the decision tree classification and regression algorithm. During the training process, control the complexity of the decision tree by setting the max_depth and min_samples_leaf parameters to prevent overfitting.

[0122] Furthermore, the edge system resource service design includes:

[0123] (A) Data Collection

[0124] According to the data type, access data is mainly divided into: time series data, structured data, and object data;

[0125] Based on the data type, different pre-processing services (time series data pre-processing, fault log pre-processing, structured data pre-processing, API interface pre-processing, etc.) are used to complete data collection;

[0126] Based on the functional deployment mode of the power plant side, the edge system of the gas turbine intelligent operation and maintenance cloud platform provides three collection methods for time series data: real-time collection, batch collection, and scheduled batch collection;

[0127] (B) Protocol Analysis

[0128] The edge system collects data from industrial control systems and provides analysis of three standard industrial communication protocols / protocols: OPC / OPCUA, ModbusTCP, and IEC60870-5-104.

[0129] In summary, the edge system platform successfully solves the bottleneck problems that may be encountered during large-scale data transmission through the use of time series databases and the implementation of customized task scheduling strategies. This not only reduces the pressure on the backbone network, but also ensures timely feedback of key data analysis results.

[0130] On the other hand, the present invention further discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor constructs the above-mentioned system.

[0131] On the other hand, the present invention further discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to construct the above-mentioned system.

[0132] In another embodiment provided in the present application, a computer program product containing instructions is also provided. When the computer is run on the computer, the computer builds a cloud-edge collaborative architecture system of any gas turbine operation and maintenance platform in the above embodiments.

[0133] It is understandable that the system provided by the embodiment of the present invention corresponds to the method provided by the embodiment of the present invention, and the explanation, examples and beneficial effects of the relevant contents can refer to the corresponding parts of the above method.

[0134] The embodiment of the present application further provides an electronic device, comprising a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus.

[0135] Memory for storing computer programs;

[0136] The processor is used to build a cloud-edge collaborative architecture system of the above-mentioned gas turbine operation and maintenance platform when executing the program stored in the memory.

[0137] The communication bus mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The communication bus can be divided into an address bus, a data bus, a control bus, etc.

[0138] The communication interface is used for communication between the above electronic device and other devices.

[0139] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage. Alternatively, the memory may be at least one storage device located remote from the processor.

[0140] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0141] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When software is used for implementation, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrations. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk).

[0142] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

[0143] In addition, it should be noted that if the embodiments of the present invention involve directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture. If the specific posture changes, the directional indications will also change accordingly.

[0144] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of such features. In addition, the meaning of "and / or" appearing throughout the text includes three parallel schemes. Taking "A and / or B" as an example, it includes scheme A, or scheme B, or schemes in which A and B are satisfied at the same time. In addition, in the embodiments of the present invention, "multiple" refers to more than two. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the ability of ordinary technicians in this field to implement. When the combination of technical solutions is mutually contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

Claims

1. A cloud-edge collaborative architecture system for a gas turbine operation and maintenance platform, characterized in that: It consists of a cloud system platform and an edge system platform, and is used to provide the following cloud-edge collaborative services: Data collaboration: After real-time data collection and processing based on the edge system platform, some data are selected to support real-time indicator updates on the front-end interface of the power plant and cloud system. Other unselected data are used to support intelligent application models for real-time computing. The model calculation results are sent to the front-end interface of the power plant and cloud system. Business application management collaboration: some application data is orchestrated in the cloud, and the other application data is sent to the edge system. Through the business application middleware and application orchestration, the application operation is calculated in real time and periodically. Service collaboration is based on the cloud system platform to provide edge applications with cloud capability call interaction, and sink cloud capabilities to the edge system platform to provide services for edge applications locally.

2. The cloud-edge collaborative architecture system of the gas turbine operation and maintenance platform according to claim 1, characterized in that: The cloud system platform consists of resource service layer IaaS, platform layer PaaS and application layer SaaS, among which: Resource service layer IaaS, which is used to provide computing, storage, network and operation and maintenance services based on basic resources to meet the business needs and different deployment methods of PaaS and SaaS layers; Platform layer PaaS, including general and professional capabilities, is used to specify service functions to meet the application development needs of the SaaS layer and the model development and training on the power plant side; Application layer SaaS is used to provide new industrial applications based on business needs and build designated business applications.

3. The cloud-edge collaborative architecture system of the gas turbine operation and maintenance platform according to claim 2, characterized in that: The computing, storage, network and operation and maintenance services of the resource service layer IaaS specifically include: Computing services: Provide multi-engine computing services including cloud hosts and containers, support application systems of different architecture types, and manage them through a unified service interface to ensure a consistent user experience; Storage service: Provides production-level cloud hard disk services and ensures data security through snapshot, full and incremental backup mechanisms; Network services: Provide virtualized network services based on software-defined network technology, provide VPC private network services, ensure strict network isolation between different projects, and provide network services including virtual routers, firewalls, public IP, and security groups. At the same time, it supports L4-L7 load balancing services to meet users' network connectivity needs in different application scenarios.

4. The cloud-edge collaborative architecture system of the gas turbine operation and maintenance platform according to claim 2, characterized in that: The overall technical architecture of the platform layer PaaS includes the following system levels: The basic layer uses container orchestration and isolation technology to isolate applications from each other and from the host operating system, ensuring that backend function upgrades are sustainable. The engine adaptation layer adapts to diversified storage computing engines to maximize resource utilization; The platform layer uses ElasticSearch to achieve high efficiency of full-text retrieval, provides efficient cache services through Redis / Mongo, improves data loading efficiency, and builds a real-time message channel through Kafka to achieve real-time data access; The service layer divides the internal gateway and external gateway into two types according to the business form, provides service authentication methods according to different business scenarios, and uses service clusters to provide stable and reliable service entrances for callers in high-concurrency scenarios. The presentation layer is encapsulated and extended based on vue and ElementUi.

5. The cloud-edge collaborative architecture system of the gas turbine operation and maintenance platform according to claim 2, characterized in that: The application layer SaaS adopts the SOA framework, takes advantage of the loose coupling between layers and the high cohesion within the layers, and realizes the componentization of business objects and unified service calls through the abstraction and mapping of the business, including: The component layer is used for component-based abstraction and lowest-granularity assembly of system business objects; The service layer integrates and highly granularly encapsulates relevant business components, and provides a standard calling interface for designated business applications for direct calling by business applications. The application layer includes specific and realizable business applications required by the system business and management levels; The system business starts at the service layer, and the transaction manager of the database is used to perform transaction control processing on the related services of the service layer.

6. The cloud-edge collaborative architecture system of the gas turbine operation and maintenance platform according to claim 1, characterized in that: The edge system platform adopts edge computing architecture, which can quickly collect, store and process data through edge nodes deployed at the edge of the network to reduce the pressure on network bandwidth; A time series database is deployed in the edge system platform to store unit data after collection, and upload the unit data to the cloud as needed to provide a sample set for training complex algorithm models on the platform side, where the trained edge algorithm service is stored in the local power plant in a lightweight mode to meet the application requirements on the power plant side; The edge algorithm service is used to open up data interaction between the edge system and the cloud system. Through real-time API services, the final calculation results of the edge system are stored in the local database and pushed to the center side regularly; The edge system platform sets customized task scheduling for massive data in an offline manner to complete the regular synchronization of massive data of the edge system to the center side; The edge system platform is internally provided with an algorithm service center, which defines different versions of services for the same model, and is used to obtain the API services published on the center side in real time, and to call them in real time to obtain the data required for participating in the edge system calculations.

7. The cloud-edge collaborative architecture system of the gas turbine operation and maintenance platform according to claim 6, characterized in that: The edge system platform includes the following system supports: The management-side application system is deployed in the production management area of ​​the power plant to form cloud-edge collaboration with the cloud system, thereby providing the power plant with intelligent applications that assist production and management; The production-side application system support, as a closed-loop feedback function module of the edge system platform, is deployed in the combined cycle control system DCS of the gas turbine, or deployed in TCS in combination with the gas turbine control system. It is used to arrange the performance optimization control server in parallel with the engineer station based on the TCS / DCS Ethernet, and deploy a reverse isolation device between the production control area and the management information area.

8. The cloud-edge collaborative architecture system of the gas turbine operation and maintenance platform according to claim 7, characterized in that: The edge algorithm service training steps of the edge system platform include: S1. Perform data preprocessing operations including noise filtering and data cleaning on the update data collected by the edge nodes according to preset requirements; S2. According to the same preset requirements, query the data records of the generator set stored in the time series database over time, and combine the query data and the data set after the preprocessing operation into a unified data set Q; S3. For data set Q, determine the amount of data in the data set. If it is greater than the preset threshold, it is considered a large-scale data set. Use batch processing technology and greedy algorithm to optimize resource data to ensure that the amount of data in the data set is within the specified threshold range. S4. Deploy the local environment, build an initial data decision tree model based on the data set, traverse the data set Q, and use the decision tree classification regression algorithm to train the algorithm model. During the training process, the max_depth and min_samples_leaf parameters are set to control the complexity of the decision tree to prevent overfitting.

9. A computer device, characterized in that: It includes a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor builds a cloud-edge collaborative architecture system of the gas turbine operation and maintenance platform as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a processor, the processor builds a cloud-edge collaborative architecture system of a gas turbine operation and maintenance platform as described in any one of claims 1 to 8.

Images