Application development system and equipment of gas turbine operation and maintenance platform and storage medium

Abstract

The invention discloses an application development system of a gas turbine operation and maintenance platform, which is characterized in that cloud-side cooperation is carried out based on a cloud system and an edge system, a thermodynamic model is taken as a core, a nonlinear equation set solver is taken as a support, and a gas turbine performance simulation analysis model is established in the platform according to equipment characteristics and system logic; and the data processing module is used for dynamically correcting the model by using real-time and historical operation data, and carrying out economic analysis, diagnosis and prediction on the equipment and the system based on the corrected accurate model. According to the method, double coupling of the thermodynamic Jacobian matrix and the data driving correction term is introduced into the nonlinear solver, so that priori knowledge and real-time data characteristics of a physical model can be fused, the limitation of a traditional solver on a high-dimension and strong-nonlinearity gas turbine system is broken through, rapid convergence and dynamic precision improvement of a simulation model are achieved, and the method is suitable for large-scale popularization and application. And the reliability of economic analysis and fault prediction is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of gas turbine data processing, and in particular to an application development system, equipment and storage medium for a gas turbine operation and maintenance platform. Background Art

[0002] Traditional gas turbine operation and maintenance systems usually rely on analysis methods based on physical models, and realize equipment status monitoring and fault diagnosis through thermodynamic models and limited data acquisition technologies. Data acquisition mostly uses centralized databases (such as relational databases) or simple time series databases to store real-time operation data. The computing framework is mainly based on offline batch processing (such as MapReduce), and the real-time performance is poor; the algorithm function is based on classical statistical methods (such as linear regression and threshold alarm), which lacks adaptability to high-dimensional and strongly nonlinear systems; the system architecture mostly adopts a monolithic design with high coupling between modules and limited scalability, making it difficult to cope with the real-time processing needs of massive heterogeneous data. Therefore, under the existing technical framework, the application development system of the gas turbine operation and maintenance platform usually faces problems such as data silos and low integration efficiency, insufficient model correction accuracy and real-time performance, weak algorithm generalization ability, and poor system collaboration.

[0003] To this end, existing technologies use ETL tools to migrate data in batches, or adopt message queues (such as Kafka) to improve transmission efficiency to achieve data integration optimization. However, the metadata conflicts and real-time bottlenecks of heterogeneous data sources cannot be fundamentally resolved during the data integration optimization process. At the same time, the conventional introduction of regularization methods (such as ridge regression) to alleviate model pathological problems, or the use of deep learning to replace physical models for solver improvement is prone to sacrificing physical interpretability and has high training costs.

[0004] In addition, existing technologies still have problems such as difficulty in balancing model convergence speed and accuracy, model separation between physical models and data-driven corrections, and existing systems relying on offline computing or hardware stacking, making it difficult to achieve efficient processing of TB-level data at low latency.

[0005] To this end, this application specifically proposes an application development system for a gas turbine operation and maintenance platform to solve the above technical problems. Summary of the Invention

[0006] The main purpose of the present invention is to provide an application development system for a gas turbine operation and maintenance platform to solve the technical problems raised in the background technology.

[0007] The present invention adopts the following technical solutions to solve the above technical problems:

[0008] An application development system for a gas turbine operation and maintenance platform is based on cloud-edge collaboration between cloud systems and edge systems, with a thermodynamic model as the core and a nonlinear equation solver as the support. A gas turbine performance simulation analysis model is established within the platform based on equipment characteristics and system logic. The system is used to dynamically correct the model using real-time and historical operating data, and to perform economic analysis, diagnosis, and prediction of equipment and systems based on the accurate model that has been corrected.

[0009] Preferably, the equation system iteration formula of the nonlinear equation system solver is:

[0010]

[0011] Among them, J k is the Jacobian matrix of the thermodynamic model, W k is the dynamic weight matrix, λ is the regularization parameter used to control the iterative stability, I is the identity matrix, D(·) represents the residual correction function based on deep learning, D h is the historical running data set, α k It is the step size controller of the physical model iteration, ensuring the stable convergence of the model under dynamic data. k It is a data-driven modified weight controller that integrates historical rules and real-time features, so α k and β k Dynamically adjust the step size coefficient and data-driven correction weight coefficient to balance solution efficiency and solution accuracy.

[0012] Preferably, based on the above nonlinear equation solver, the system includes:

[0013] The data acquisition module directly obtains managed data from the power plant's existing data platform, or collects data from on-site business systems through a distributed heterogeneous data integration engine, and completes cleaning and storage.

[0014] The data computing module is used for the real-time computing engine and the distributed offline computing framework, and processes streaming data and batch data respectively to provide real-time monitoring and offline analysis capabilities;

[0015] The data management module is used to perform data management operations including metadata management, data standards management, data quality management, and tag management to form a global data asset map;

[0016] The algorithm function module integrates specified machine learning algorithms and industrial models to support model development, training, optimization, and service-oriented closed loop;

[0017] Developer function module, used to provide microservice architecture, container services and DevOps tool chain.

[0018] Preferably, in the data acquisition module:

[0019] When a power plant already has a data platform, it can implement streaming data transmission through KafkaConnect and complete metadata management, task scheduling, and data reconciliation based on a distributed heterogeneous data integration engine.

[0020] When a power plant does not have a data platform, data is collected directly from the PI database and maintenance system through a star connection, and hierarchical storage is adopted in a time series database and a relational database. The time series database supports writing millions of data points in seconds and querying TB-level data in milliseconds, and the relational database implements distributed storage of massive data based on HBase and HDFS.

[0021] Preferably, in the data calculation module:

[0022] The real-time computing engine uses Flink to process streaming data and simplifies development through SQL code;

[0023] The offline computing framework is based on Spark and Hive engines, supports HiveQL language and Tez execution engine, and is compatible with multiple data storage formats;

[0024] The results of the real-time calculation and offline calculation are dynamically fed back to the front end through a visual interface for decision-making.

[0025] Preferably, in the data management module:

[0026] Metadata management records data processing relationships, forming a three-level data map of power plants, groups, and industries. Data quality management uses verification rules and re-inspection mechanisms to trace the source of quality issues and accumulate a governance knowledge base.

[0027] Data tag management implements full life cycle management of tags and supports multi-dimensional data classification and retrieval.

[0028] Preferably, in the algorithm function module:

[0029] Integrate clustering, classification, decision tree, neural network and specified similarity calculation algorithms;

[0030] Supports automatic feedback and secondary training of model inference results to achieve continuous iterative optimization of the algorithm;

[0031] Based on the training and inference computing framework, a scenario-based algorithm system is developed to meet the needs of gas turbine early warning and diagnosis.

[0032] Preferably, in the developer function module:

[0033] The microservice architecture implements service registration, configuration center and distributed transaction management based on SpringCloud;

[0034] Container Service is based on Kubernetes and supports the full lifecycle management of stateless, stateful, and scheduled tasks.

[0035] Multi-tenant management ensures system security and resource utilization through resource quota allocation and permission level isolation.

[0036] Preferably, in the API service module:

[0037] Centrally display service interfaces through the API marketplace, supporting on-demand application and rapid generation;

[0038] The API gateway implements service security management and traffic monitoring, and supports the standardized opening of data sets, labels, and algorithm services;

[0039] The data service bus adopts the REST API architecture and is integrated with the enterprise authentication system to ensure that interface call permissions are controllable.

[0040] In another aspect, the present invention further discloses a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the steps of the above method.

[0041] On the other hand, the present invention further discloses a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the above method.

[0042] It can be seen from the above technical solutions that the present invention provides an application development system for a gas turbine operation and maintenance platform.

[0043] Compared with the prior art, the present invention has the following advantages:

[0044] 1. By introducing the dual coupling of the thermodynamic Jacobian matrix and the data-driven correction term into the nonlinear solver, the present invention can integrate the prior knowledge of the physical model with the real-time data characteristics, thereby overcoming the limitations of traditional solvers for high-dimensional and strongly nonlinear combustion turbine systems, achieving rapid convergence of the simulation model and improving the dynamic accuracy, thereby enhancing the reliability of economic analysis and fault prediction.

[0045] 2. The present invention sets up a cloud-edge collaborative architecture and a heterogeneous data integration engine for the cloud and edge systems in the data acquisition module, which can integrate distributed heterogeneous data sources, eliminate data silos and achieve efficient data integration, and provide comprehensive and real-time data support for the gas turbine performance simulation and analysis model.

[0046] 3. The present invention adopts a time series database and a distributed relational database in the data storage layer, which supports high-speed writing and query of massive data while meeting the dual needs of real-time analysis and historical backtracking, thereby improving the processing efficiency and storage reliability of gas turbine operation data.

[0047] 4. By integrating distributed computing engines such as Spark and Flink and the Tez execution engine in the data computing module, the present invention can support offline batch processing and real-time streaming processing, thereby improving data processing speed and model dynamic correction capabilities, and achieving rapid response to gas turbine economic analysis and fault warning.

[0048] 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

[0049] 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:

[0050] Figure 1 Schematic diagram of the overall system structure of the present invention;

[0051] Figure 2 This is a schematic diagram of the structure of the gas turbine operation and maintenance cloud platform system of the present invention;

[0052] Figure 3 This is a schematic diagram of the data standard management system architecture of the present invention;

[0053] Figure 4 Schematic diagram of the data quality management system architecture of the present invention;

[0054] Figure 5 This is a schematic diagram of the data tag management system architecture of the present invention;

[0055] Figure 6 This is a schematic diagram of the algorithm function module system structure of the present invention. DETAILED DESCRIPTION

[0056] 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.

[0057] In the embodiment, see Figures 1 to 6 .

[0058] like Figure 1 As shown, an application development system for a gas turbine operation and maintenance platform proposed in an embodiment of the present invention is based on cloud-edge collaboration between cloud systems and edge systems, and takes thermodynamic models as the core and nonlinear equation solvers as the support. A gas turbine performance simulation analysis model is established within the platform according to equipment characteristics and system logic, which is used to dynamically modify the model using real-time and historical operating data, and to perform economic analysis, diagnosis, and prediction of equipment and systems based on the accurate model after the modification. At this time, the system structure of the gas turbine operation and maintenance platform is as follows: Figure 2 As shown, the iterative formula of the nonlinear equation solver is:

[0059]

[0060] Where, J k is the Jacobian matrix of the thermodynamic model, W k is a dynamic weight matrix, which is adjusted based on real-time data quality (such as data cleaning rules and quality inspection results) to achieve dynamic model correction. λ is a regularization parameter used to control iterative stability. I is the unit matrix. D(·) represents the residual correction function based on deep learning. D h For historical running data sets, quickly extract features through time series database, α k It is the step size controller of the physical model iteration, ensuring the stable convergence of the model under dynamic data. k It is a data-driven modified weight controller that integrates historical rules and real-time features, so α k and β k Dynamically adjust the step size coefficient and data-driven correction weight coefficient to balance solution efficiency and solution accuracy.

[0061] At this time, by introducing the dual coupling of thermodynamic Jacobian matrix and data-driven correction term in the nonlinear solver, it is possible to integrate the prior knowledge of the physical model and the real-time data characteristics, so as to break through the limitations of traditional solvers for high-dimensional and strongly nonlinear gas turbine systems, achieve rapid convergence and dynamic accuracy improvement of the simulation model, and improve the reliability of economic analysis and fault prediction. At the same time, by setting up the cloud-edge collaborative architecture and heterogeneous data integration engine of the cloud and edge systems in the data acquisition module, it is possible to integrate distributed heterogeneous data sources, eliminate data silos and realize efficient data integration, and provide comprehensive and real-time data support for the gas turbine performance simulation analysis model.

[0062] Based on the above nonlinear equation solver, the system includes:

[0063] (1) Data acquisition module, which directly obtains the processed data from the power plant's existing data platform or collects data from the on-site business system through a distributed heterogeneous data integration engine, and completes cleaning and storage.

[0064] When a power plant already has a data platform, it can implement streaming data transmission through KafkaConnect and complete metadata management, task scheduling, and data reconciliation based on a distributed heterogeneous data integration engine.

[0065] When a power plant does not have a data platform, data is collected directly from the PI database and maintenance system through a star connection, and hierarchical storage is adopted in a time series database and a relational database. The time series database supports writing millions of data points in seconds and querying TB-level data in milliseconds, and the relational database implements distributed storage of massive data based on HBase and HDFS.

[0066] It should be noted that the data is stored in the database, including:

[0067] A) Time Series Database: The gas turbine intelligent operation and maintenance cloud platform integrates a high-performance, lightweight time series data management system for the Industrial Internet of Things. It features low storage costs, high-speed data writing (writing millions of data points in seconds), fast query (querying TB-level data in milliseconds), data addition, deletion, modification, and query, rich aggregation functions, similarity matching, integrated query and analysis (one piece of data can meet real-time query and analysis mining), and standard interfaces (using standard JDBC interfaces and SQL-like query languages).

[0068] B) Relational database: The gas turbine intelligent operation and maintenance cloud platform integrates distributed databases including online transaction processing (OLTP) and online analytical processing (OLAP).

[0069] In addition, the gas turbine intelligent operation and maintenance cloud platform is based on Apache HBase, uses Hadoop HDFS as the file storage system, uses Hadoop MapReduce to process massive data, and uses Zookeeper as a collaborative service to solve the theoretical and implementation limitations of relational databases when processing massive data, and realizes massive data storage and high-speed reading and writing.

[0070] In addition, data collection adopts a star connection, which aggregates from the edge side to the production side without generating horizontal transmission, thus avoiding data diffusion; at the same time, during the transmission and synchronization process, it is transmitted through a dedicated line to ensure that data is not leaked and data security is guaranteed.

[0071] For production-side data storage, we use the distributed file system HDFS. HDFS's inherent HA redundancy architecture ensures that data in big data storage is not lost, ensuring the security and stability of big data storage itself. The underlying data storage uses multiple copies to avoid single points of failure.

[0072] In summary, by using a time-series database and a distributed relational database in the data storage layer, we can support high-speed writing and querying of massive data while meeting the dual needs of real-time analysis and historical backtracking, thereby improving the processing efficiency and storage reliability of gas turbine operating data.

[0073] (2) Data computing module, which is used for the real-time computing engine and the distributed offline computing framework, and processes streaming data and batch data respectively to provide real-time monitoring and offline analysis functions, including:

[0074] The real-time computing engine uses Flink to process streaming data and simplifies development through SQL code;

[0075] The offline computing framework is based on Spark and Hive engines, supports HiveQL language and Tez execution engine, and is compatible with multiple data storage formats;

[0076] The results of real-time and offline calculations are dynamically fed back to the front end through a visual interface for decision-making.

[0077] Further, adopt:

[0078] A) Distributed Offline Computing Framework

[0079] The offline computing of the gas turbine intelligent operation and maintenance cloud platform includes big data offline data development, task management, operation and maintenance management, monitoring and early warning. It supports visual management and control of the big data processing process, shields the complex underlying distributed computing engine, and conducts online data development based on the Web to improve development efficiency.

[0080] The data storage layer primarily addresses the problem of storing offline data and currently uses the HDFS (Distributed File System) for storage. The computing engine uses the Spark and Hive engines to process offline computing task scripts. Data processing primarily includes offline data processing task scripts, task scheduling, and task operation and maintenance. The visual development interface is primarily for offline data development engineers, providing a visual task development interface.

[0081] The gas turbine intelligent operation and maintenance cloud platform includes offline services such as batch processing and data warehousing. It can perform batch analysis and aggregation of structured and semi-structured data to complete data calculations. It also provides the SQL-like Hive Query Language to operate on structured data, allowing users familiar with SQL to query and analyze data.

[0082] The gas turbine intelligent operation and maintenance cloud platform uses the new execution engine Tez instead of conventional MapReduce, supporting multiple data storage formats such as JSON, CSV, TEXTFILE, RCFILE, SEQUENCEFILE, ORC, etc. It also supports in-memory computing engines such as Spark and Presto to speed up queries.

[0083] B) Real-time calculation

[0084] The gas turbine intelligent operation and maintenance cloud platform provides a one-stop development platform for real-time data collection, calculation, monitoring and early warning. It supports the processing of streaming data through SQL code, simplifies the learning threshold of streaming computing for developers, and provides a web-based visual operation interface to improve the efficiency of streaming data development.

[0085] Data sources primarily consist of Oracle, MySQL, and business system logs. Data is collected from these data sources using collection engines such as Kafka, Flume, and Logstash, and then stored in Kafka. The Flink real-time computing engine processes real-time computing tasks. Data development provides maintenance and management for real-time collection and computing tasks, with monitoring and early warning capabilities providing visibility into task execution. A visual development interface is provided for real-time computing development engineers to improve development efficiency.

[0086] Real-time data collection is customized and developed based on open source components, supporting real-time collection of mainstream relational databases such as Oracle, MySQL, SQLServer, as well as domestic relational databases such as DAMO.

[0087] In summary, by integrating distributed computing engines such as Spark and Flink and the Tez execution engine in the data computing module, it is possible to support offline batch processing and real-time streaming processing, thereby improving data processing speed and model dynamic correction capabilities, and achieving rapid response to gas turbine economic analysis and fault warning.

[0088] (3) Data management module, which is used to perform data management operations including metadata management, data standard management, data quality management and label management to form a global data asset map. Figure 3 、 Figure 4 and Figure 5 ,in:

[0089] Metadata management records the data processing lineage relationship, forming a three-level data map of power plant, group, and industry. Specifically, it can uniformly manage various metadata of the data warehouse and fully record the lineage relationship of the data processing link, forming a global data map of the power plant, group, and industry gas turbine, facilitating data management and application at all levels. At this time, by setting metadata management, data standard management, and quality inspection functions in the data management module, it can play a role in standardizing data lineage relationships and governance processes, thereby ensuring data integrity and accuracy, and ultimately providing a high-quality data asset foundation for gas turbine diagnosis and prediction;

[0090] Data standards management is used to cleanse and process data according to established standards, improve data standardization, and form preliminary data assets. Key functions include online definition of data standards, approval standards, standard-based modeling, definition of standardization processing rules, and configuration of standardization processing tasks.

[0091] Data quality management traces the source of quality issues through verification rules and re-inspection mechanisms, and accumulates a governance knowledge base. This allows for verification based on quality rules, checking for quality issues in platform tables and fields, and effectively tracing back to the source of quality issues. The data quality management module continuously generates positive feedback through mechanisms such as re-inspection operations, effectively accumulating data quality governance knowledge and providing support and assurance for the platform's subsequent data rules.

[0092] Data tag management implements full life cycle management of tags and supports multi-dimensional data classification and retrieval.

[0093] (4) Algorithm function module, which integrates specified machine learning algorithms and industrial models to support model development, training, optimization and service closed loop, including:

[0094] Integrate clustering, classification, decision tree, neural network and specified similarity calculation algorithms;

[0095] Supports automatic feedback and secondary training of model inference results to achieve continuous iterative optimization of the algorithm;

[0096] Based on the training and inference computing framework, a scenario-based algorithm system is developed to meet the needs of gas turbine early warning and diagnosis.

[0097] Further, such as Figure 6 As shown in the figure, this algorithm function module further enhances modeling data processing capabilities based on the IaaS layer and general PaaS functions. It implements the full process of algorithm model development, training, management, and service operation. By integrating industrial algorithms such as clustering and neural networks and model closed-loop optimization mechanisms into the algorithm function module, it can support the development of multi-scenario gas turbine diagnostic models, thereby improving the algorithm's adaptive iteration capabilities and ultimately achieving accurate prediction and continuous optimization of gas turbine operation and maintenance.

[0098] (5) Developer function module, which is used to provide microservice architecture, container services and DevOps tool chain, including:

[0099] The microservice architecture implements service registration, configuration center and distributed transaction management based on SpringCloud;

[0100] Container Service is based on Kubernetes and supports the full lifecycle management of stateless, stateful, and scheduled tasks.

[0101] Multi-tenant management ensures system security and resource utilization through resource quota allocation and permission level isolation.

[0102] In summary, by providing a microservice framework, containerized deployment, and DevOps tool chain in the developer function module, it can lower the development threshold and improve the scalability of the system, thereby supporting multi-team collaboration and rapid application iteration, and finally meeting the efficient operation and maintenance and customization needs of the gas turbine operation and maintenance platform.

[0103] (6) API service module, which displays service interfaces in a centralized manner through the API market, supports on-demand application and rapid generation, including:

[0104] The API gateway implements service security management and traffic monitoring, and supports the standardized opening of data sets, labels, and algorithm services;

[0105] The data service bus adopts the REST API architecture and is integrated with the enterprise authentication system to ensure that interface call permissions are controllable.

[0106] In summary, by building a unified data service bus and API marketplace in the API service module, we can open up the last mile between data and applications, thereby realizing the rapid service of data assets and seamless integration with third-party systems, and finally enhancing the platform's openness and adaptability to business scenarios. At the same time, by adopting dedicated line transmission, multi-copy storage and security group management in the full-link design, we can ensure the security of data transmission and storage, thereby avoiding single point failures and data leakage risks, and finally ensuring the stable operation and compliance of the gas turbine operation and maintenance system.

[0107] In another aspect, the present invention further discloses a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the steps of the above method.

[0108] On the other hand, the present invention further discloses a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the above method.

[0109] In another embodiment provided in the present application, a computer program product including instructions is also provided, which, when executed on a computer, enables the computer to execute the application development system of any gas turbine operation and maintenance platform in the above embodiments.

[0110] 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.

[0111] 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.

[0112] Memory for storing computer programs;

[0113] The processor is used to implement the application development system of the above-mentioned gas turbine operation and maintenance platform when executing the program stored in the memory.

[0114] The communication bus mentioned in the above electronic device can be a peripheral component interconnect standard bus or an extended industry standard architecture bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc.

[0115] The communication interface is used for communication between the above electronic device and other devices.

[0116] The memory may include a random access memory, or a non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0117] The above-mentioned processor can be a general-purpose processor, including a central processing unit, a network processor, etc.; it can also be a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component.

[0118] 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 by wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. 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, or a semiconductor medium (e.g., a solid-state hard disk).

[0119] 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.

[0120] 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.

[0121] 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. An application development system for a gas turbine operation and maintenance platform, characterized in that: Based on cloud-edge collaboration between cloud systems and edge systems, with thermodynamic models as the core and supported by nonlinear equation solvers, a gas turbine performance simulation analysis model is established within the platform according to equipment characteristics and system logic. The model is dynamically modified using real-time and historical operating data, and the economic analysis, diagnosis and prediction of equipment and systems are carried out based on the accurate model after the modification.

2. The application development system of the gas turbine operation and maintenance platform according to claim 1, characterized in that: The equations iteration formula of the nonlinear equations solver is: Among them, J k is the Jacobian matrix of the thermodynamic model, W k is the dynamic weight matrix, λ is the regularization parameter used to control the iterative stability, I is the identity matrix, D(·) represents the residual correction function based on deep learning, D h is the historical running data set, α k and β k Dynamically adjust the step size coefficient and data-driven correction weight coefficient to balance solution efficiency and solution accuracy.

3. The application development system of the gas turbine operation and maintenance platform according to claim 2, characterized in that: include: The data acquisition module directly obtains managed data from the power plant's existing data platform, or collects data from on-site business systems through a distributed heterogeneous data integration engine, and completes cleaning and storage. The data computing module is used for the real-time computing engine and the distributed offline computing framework, and processes streaming data and batch data respectively to provide real-time monitoring and offline analysis capabilities; The data management module is used to perform data management operations including metadata management, data standards management, data quality management, and tag management to form a global data asset map; The algorithm function module integrates specified machine learning algorithms and industrial models to support model development, training, optimization, and service-oriented closed loop; Developer function module, used to provide microservice architecture, container services and DevOps tool chain.

4. The application development system of the gas turbine operation and maintenance platform according to claim 3, characterized in that: In the data acquisition module: When a power plant already has a data platform, it can implement streaming data transmission through KafkaConnect and complete metadata management, task scheduling, and data reconciliation based on a distributed heterogeneous data integration engine. When a power plant does not have a data platform, data is collected directly from the PI database and maintenance system through a star connection, and hierarchical storage is adopted in a time series database and a relational database. The time series database supports writing millions of data points in seconds and querying TB-level data in milliseconds, and the relational database implements distributed storage of massive data based on HBase and HDFS.

5. The application development system of the gas turbine operation and maintenance platform according to claim 3, characterized in that: In the data calculation module: The real-time computing engine uses Flink to process streaming data and simplifies development through SQL code; The offline computing framework is based on Spark and Hive engines, supports HiveQL language and Tez execution engine, and is compatible with multiple data storage formats; The results of the real-time calculation and offline calculation are dynamically fed back to the front end through a visual interface for decision-making.

6. The application development system of the gas turbine operation and maintenance platform according to claim 3, characterized in that: In the data management module: Metadata management records the data processing relationship to form a data map; Data quality management traces the source of quality issues through verification rules and re-inspection mechanisms, and accumulates a governance knowledge base; Data tag management implements full life cycle management of tags and supports multi-dimensional data classification and retrieval.

7. The application development system of the gas turbine operation and maintenance platform according to claim 3, characterized in that: In the algorithm function module: Integrate clustering, classification, decision tree, neural network and specified similarity calculation algorithms; Supports automatic feedback and secondary training of model inference results to achieve continuous iterative optimization of the algorithm; Based on the training and inference computing framework, a scenario-based algorithm system is developed to meet the needs of gas turbine early warning and diagnosis.

8. The application development system of the gas turbine operation and maintenance platform according to claim 3, characterized in that: In the developer function module: The microservice architecture implements service registration, configuration center and distributed transaction management based on SpringCloud; Container Service is based on Kubernetes and supports the full lifecycle management of stateless, stateful, and scheduled tasks. Multi-tenant management ensures system security and resource utilization through resource quota allocation and permission level isolation.

9. 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 is caused to perform the steps of the method according to any one of claims 1 to 8.

10. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method according to any one of claims 1 to 8.

Images