A power station data fusion processing system and method based on hyper-converged architecture

Abstract

The application discloses a power station data fusion processing system and method based on a super-fusion architecture, relates to the technical field of power station data fusion processing systems, and comprises a super-fusion infrastructure layer, a multi-source data acquisition layer, a data preprocessing layer, a layered data fusion layer, an intelligent analysis layer, a security protection layer, a unified management platform layer, an edge collaboration layer and an application output layer, each layer is realized through a software-defined network to realize bidirectional data interaction and resource collaboration, and the super-fusion infrastructure layer provides integrated computing power, storage and network resource support for the whole system. Through a three-level fusion architecture of data-features-decision, the pain point of a shallow fusion level caused by the data split of traditional devices is solved, multi-source data is synchronized at a microsecond level, the fusion result is verified in multiple dimensions, the fault diagnosis accuracy is improved, the prediction error is reduced, and accurate data support is provided for power station operation and maintenance.

Description

Technical Field

[0001] This invention relates to the field of power plant data fusion processing system technology, and specifically to a power plant data fusion processing system and method based on a hyper-converged architecture. Background Technology

[0002] As a core facility for energy supply, the accurate monitoring and intelligent decision-making of the power plant's operating status rely on the efficient integration and processing of multi-source data. Data processing must meet the requirements of real-time performance, reliability, and security; otherwise, data fragmentation and processing delays may lead to missed equipment failures, loss of power generation efficiency, and even grid dispatch risks.

[0003] Traditional power plant data processing systems generally adopt a distributed hardware and monolithic software architecture. The core pain point is that the data fusion layer is shallow and lacks a full-link layered fusion design of "data-feature-decision". The inconsistent protocols of heterogeneous acquisition equipment on site and the spatiotemporal asynchrony of multi-source data make it impossible to effectively coordinate and deeply analyze massive monitoring data. The fusion results are not accurate enough and are prone to causing missed or misjudged equipment faults, continuous loss of power generation efficiency, and in severe cases, it can also cause safety risks such as grid dispatch instability. Summary of the Invention

[0004] The purpose of this invention is to provide a power plant data fusion processing system and method based on a hyper-converged architecture, which solves the problem that traditional power plant data processing systems generally adopt a distributed hardware plus a single software architecture, resulting in shallow data fusion layers and a lack of a full-link layered fusion design of "data-feature-decision".

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A power plant data fusion processing system based on a hyperconverged architecture includes a hyperconverged infrastructure layer, a multi-source data acquisition layer, a data preprocessing layer, a hierarchical data fusion layer, an intelligent analysis layer, a security protection layer, a unified management platform layer, an edge collaboration layer, and an application output layer. Each layer achieves bidirectional data interaction and resource collaboration through a software-defined network. The hyperconverged infrastructure layer provides integrated computing power, storage, and network resource support for the entire system.

[0006] A further improvement of this invention is that the hyperconverged infrastructure layer adopts a distributed architecture, integrating a virtualization engine, a distributed storage module, and a software-defined network controller; the virtualization engine supports KVM or VMware vSphere architecture, abstracting physical server resources into a virtual resource pool to achieve dynamic allocation and elastic expansion of CPU and memory resources; the distributed storage module uses a combination of replication mechanism and EC erasure coding to store power plant data, sharding the data and storing it on multiple nodes, while configuring a cold and hot data intelligent scheduling strategy, deploying high-frequency access data on SSD storage media and low-frequency data on HDD storage media.

[0007] A further improvement of the present invention is that the multi-source data acquisition layer includes a sensor cluster, a protocol conversion module, and a data acquisition terminal. The sensor cluster covers the power plant system and transmission lines, and includes vibration sensors, temperature sensors, pressure sensors, and infrared array detectors. The protocol conversion module has a built-in OPCUA adapter to realize unified conversion between different protocols such as Modbus / TCP and MQTT, and is compatible with data interaction between legacy PLC controllers and new intelligent devices.

[0008] A further improvement of this invention is that the data preprocessing layer includes a data cleaning module, a spatiotemporal alignment module, and a standardization module. The data cleaning module uses a generative adversarial network to correct sensor drift data, completes missing data through trend extrapolation, and removes invalid noise data. The spatiotemporal alignment module is based on the IEEE 1588v2 precision clock protocol and combines Kalman filtering to compensate for transmission delay, achieving microsecond-level synchronization between 10kHz sampling of vibration signals and 1Hz sampling of thermal parameters. The standardization module unifies the dimensions and formats of multi-source data and generates a standardized dataset that meets the requirements of fusion processing.

[0009] A further improvement of this invention is that the hierarchical data fusion layer includes a data layer fusion module, a feature layer fusion module, and a decision layer fusion module. The data layer fusion module uses a weighted average algorithm to fuse multiple sets of data from the same source for the same monitoring object, preserving the details of the original data. The feature layer fusion module extracts time-domain features, frequency-domain features, and time-frequency features from the data, filters key degradation indicators using a random forest algorithm, and then simplifies the data dimensions using a principal component analysis algorithm. The decision layer fusion module, based on DS evidence theory and Bayesian networks, integrates the preliminary decision results from various data sources to achieve multi-dimensional verification of fault diagnosis and risk assessment.

[0010] A further improvement of this invention is that the intelligent analysis layer includes a digital twin modeling module, a remaining useful life prediction module, and a dynamic risk assessment module; the digital twin modeling module adopts multiphysics coupling simulation technology to construct thermal, mechanical, and control models to achieve high-fidelity simulation of the operating status of power plant equipment; the RUL prediction module adopts a hybrid framework that integrates LSTM networks and physical degradation models; and the dynamic risk assessment module constructs a risk matrix to quantify the probability of failure and the severity of its consequences, guiding the prioritization of maintenance.

[0011] A further improvement of this invention is that the security protection layer conforms to the IEC62443 standard and includes an access control module, a data encryption module, and an operation audit module. The access control module adopts a role-based access control mechanism to restrict the system operation scope of different maintenance personnel. The data encryption module uses SSL / TLS encryption for transmitted data and AES-256 encryption for stored data, while using snapshot and continuous data protection technologies to achieve real-time data backup and fault rollback. The operation audit module records all system operation behaviors, generates traceable audit logs, and automatically triggers alarms for abnormal operations.

[0012] A further improvement of this invention is that the unified management platform layer integrates a panoramic monitoring module, an automated operation and maintenance module, and a visual interaction module; the panoramic monitoring module displays the CPU, memory, disk usage, and network bandwidth of each node in real time, and automatically triggers tiered alarms when equipment operating parameters are abnormal; the automated operation and maintenance module supports scripted task configuration, enabling automated operations such as regular database backups and equipment status inspections; the visual interaction module constructs a 3D visualization interface based on UnityHDRP pipelines, displaying equipment health heatmaps, pipeline corrosion profiles, and AR inspection navigation annotations, and supports gesture-based virtual dashboard adjustments for simulation parameters.

[0013] A further improvement of this invention is that the edge collaboration layer deploys an industrial-grade edge computing gateway to achieve local preprocessing and real-time analysis of power plant field data, uploading only key fusion results and abnormal data to the cloud, thereby reducing data transmission pressure and latency; the application output layer includes an equipment operation and maintenance module, a power generation optimization module, and a scheduling decision module. The equipment operation and maintenance module outputs predictive maintenance work orders, the power generation optimization module optimizes unit operating parameters based on fused data, and the scheduling decision module provides accurate data support for power grid scheduling.

[0014] A power plant data fusion processing method based on a hyperconverged architecture, the method being based on the aforementioned power plant data fusion processing system based on a hyperconverged architecture, comprising: The hyperconverged infrastructure layer first integrates and dynamically allocates computing power, storage, and network resources; the multi-source data acquisition layer collects multi-dimensional monitoring data from power plant equipment and transmission lines and completes unified protocol conversion; the data preprocessing layer cleans, aligns, and standardizes the collected data to generate a standardized dataset; the layered data fusion layer sequentially performs three-level fusion processing of the data layer, feature layer, and decision layer to extract the core value of the data; the intelligent analysis layer completes digital twin modeling, remaining service life prediction, and dynamic risk assessment based on the fused data; the security protection layer implements access control, data encryption, and operation auditing throughout the process to ensure system and data security; the unified management platform layer realizes panoramic system monitoring, automated operation and maintenance, and three-dimensional visualization interactive management; the edge collaboration layer completes local preprocessing and real-time analysis of on-site data, uploading only key data to the cloud; and the application output layer outputs relevant instructions and data for equipment operation and maintenance, power generation optimization, and scheduling decisions based on the processing and analysis results to support the efficient and stable operation of the power plant.

[0015] Compared with the prior art, the present invention has at least the following beneficial technical effects: This invention addresses the pain point of shallow fusion caused by fragmented data in traditional devices through a three-level data-feature-decision fusion architecture. It achieves microsecond-level synchronization of multi-source data, and the fusion results are verified across multiple dimensions, improving fault diagnosis accuracy and reducing prediction errors, thus providing precise data support for power plant operation and maintenance. Furthermore, this invention integrates computing power, storage, and network resources, dynamically allocating resources through a virtual resource pool, thereby improving resource utilization. Simultaneously, it utilizes edge collaboration and hot / cold data scheduling to reduce data transmission latency and improve processing efficiency compared to traditional distributed architectures, adapting to the real-time processing needs of massive amounts of data in power plants.

[0016] Furthermore, by adhering to the IEC62443 standard, this invention employs end-to-end encryption, hierarchical access control, and operation auditing to reduce the risk of data leakage and unauthorized access. Simultaneously, by utilizing automated operation and maintenance and 3D visualization interaction, it improves the efficiency of operation and maintenance work order processing and reduces overall operation and maintenance costs, making it suitable for various intelligent management and control scenarios for power plants. Attached Figure Description

[0017] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the overall architecture of the power plant data fusion processing system based on a hyper-converged architecture according to the present invention. Figure 2This is a flowchart of the multi-source data acquisition and preprocessing process of the present invention; Figure 3 This is a flowchart of the hierarchical data fusion and intelligent analysis process of the present invention; Figure 4 This is a flowchart illustrating the security protection and unified management process of the present invention. Figure 5 This is a flowchart illustrating the edge collaboration and application output of the present invention. Detailed Implementation

[0019] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the spirit or scope of the invention. Therefore, the drawings and description are considered to be exemplary in nature and not restrictive.

[0020] In the description of this invention, it should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0021] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0022] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0023] The accompanying drawings illustrate various structural schematic diagrams according to embodiments disclosed in this invention. These drawings are not to scale, and some details have been enlarged for clarity, and some details may have been omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.

[0024] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0025] Example 1 like Figure 1-5As shown in the figure, this embodiment provides a power plant data fusion processing system based on a hyperconverged architecture, including a hyperconverged infrastructure layer, a multi-source data acquisition layer, a data preprocessing layer, a hierarchical data fusion layer, an intelligent analysis layer, a security protection layer, a unified management platform layer, an edge collaboration layer, and an application output layer. Each layer realizes bidirectional data interaction and resource collaboration through a software-defined network. The hyperconverged infrastructure layer provides integrated computing power, storage, and network resource support for the entire system.

[0026] like Figure 1-3 As shown, the hyperconverged infrastructure layer adopts a distributed architecture, integrating a virtualization engine, a distributed storage module, and a software-defined network controller. The virtualization engine supports KVM or VMware vSphere architecture, abstracting physical server resources into a virtual resource pool to achieve dynamic allocation and elastic scaling of CPU and memory resources. The distributed storage module uses a combination of replication mechanism and EC erasure coding to store power plant data, sharding the data and storing it on multiple nodes. It also configures a smart scheduling strategy for hot and cold data, deploying high-frequency access data on SSD storage media and low-frequency data on HDD storage media.

[0027] The virtualization engine abstracts physical servers into virtual resource pools, dynamically allocating CPU and memory resources, improving resource utilization, and supporting elastic scaling during peak business periods. Distributed storage uses a combination of replication and EC erasure coding, with data sharded and stored on multiple nodes. Combined with hot and cold data scheduling, high-frequency data is stored on SSDs and low-frequency data is stored on HDDs, improving storage reliability and reducing access latency. Software-defined networking enables seamless data interaction at all levels, eliminating the risk of single points of failure. The expansion cycle is shortened from weeks to hours, significantly reducing deployment and maintenance costs.

[0028] like Figure 1 and Figure 3 As shown, the multi-source data acquisition layer includes a sensor cluster, a protocol conversion module, and a data acquisition terminal. The sensor cluster covers the power plant system and transmission lines, including vibration sensors, temperature sensors, pressure sensors, and infrared array detectors. The protocol conversion module has a built-in OPCUA adapter to achieve unified conversion of different protocols such as Modbus / TCP and MQTT, and is compatible with data interaction between legacy PLC controllers and new intelligent devices. The data preprocessing layer includes a data cleaning module, a spatiotemporal alignment module, and a standardization module. The data cleaning module uses a generative adversarial network to correct sensor drift data, completes missing data through trend extrapolation, and removes invalid noise data. The spatiotemporal alignment module is based on the IEEE 1588v2 precision clock protocol and combines Kalman filtering to compensate for transmission delay, achieving microsecond-level synchronization of 10kHz sampling of vibration signals and 1Hz sampling of thermal parameters. The standardization module unifies the dimensions and formats of multi-source data and generates a standardized dataset that meets the requirements of fusion processing.

[0029] The sensor cluster covers power plant equipment and transmission lines, collecting multi-dimensional data such as vibration and temperature. The protocol conversion module uses an OPCUA adapter to unify protocols such as Modbus / TCP and MQTT. The data cleaning module uses generative adversarial networks to correct drifting data and trend extrapolation to complete missing data. The spatiotemporal alignment module is based on the IEEE1588v2 protocol and Kalman filtering to achieve microsecond-level synchronization. The standardization module unifies the units and formats and outputs standardized data. After preprocessing, the invalid noise removal rate and the synchronization accuracy of multi-source data are improved, providing reliable data support for subsequent fusion analysis.

[0030] like Figure 1 and Figure 3 As shown, the hierarchical data fusion layer includes a data layer fusion module, a feature layer fusion module, and a decision layer fusion module. The data layer fusion module uses a weighted average algorithm to fuse multiple sets of data from the same source for the same monitoring object, preserving the details of the original data. The feature layer fusion module extracts time-domain features, frequency-domain features, and time-frequency features from the data, filters key degradation indicators using a random forest algorithm, and then simplifies the data dimensions using a principal component analysis algorithm. The decision layer fusion module, based on DS evidence theory and Bayesian networks, integrates preliminary decision results from various data sources to achieve multi-dimensional verification of fault diagnosis and risk assessment. The intelligent analysis layer includes a digital twin modeling module, a remaining useful life prediction module, and a dynamic risk assessment module. The digital twin modeling module uses multi-physics coupling simulation technology to construct thermal, mechanical, and control models to achieve high-fidelity simulation of the operating status of power plant equipment. The RUL prediction module uses a hybrid framework that integrates LSTM networks and physical degradation models. The dynamic risk assessment module constructs a risk matrix to quantify the probability of fault occurrence and the severity of consequences, guiding the prioritization of maintenance.

[0031] The data layer uses a weighted average algorithm to fuse data from the same source, preserving original details; the feature layer extracts time-domain, frequency-domain, and time-frequency features, selects key indicators through random forest, and simplifies dimensions through principal component analysis; the decision layer integrates multi-source decision results based on DS evidence theory and Bayesian networks to improve the credibility of fault diagnosis; the digital twin modeling module achieves high-fidelity simulation of equipment operation through multi-physics coupling simulation; the RUL prediction module uses an LSTM network + physical degradation model to improve prediction accuracy; and the dynamic risk assessment module quantifies the probability and consequences of failure to guide maintenance priority ranking.

[0032] like Figure 1 and Figure 5As shown, the security protection layer conforms to the IEC62443 standard and includes an access control module, a data encryption module, and an operation audit module. The access control module adopts a role-based access control mechanism to restrict the system operation scope of different maintenance personnel. The data encryption module uses SSL / TLS encryption for transmitted data and AES-256 encryption for stored data. It also uses snapshot and continuous data protection technologies to achieve real-time data backup and fault rollback. The operation audit module records all system operation behaviors, generates traceable audit logs, and automatically triggers alarms for abnormal operations.

[0033] Access control employs role-based access control to restrict the scope of operations; transmitted data is encrypted with SSL / TLS, and stored data is encrypted with AES-256, combined with snapshots and continuous data protection to enable fault rollback; operation auditing records all behaviors, and abnormal operations are automatically alerted to prevent unauthorized access and data leakage.

[0034] like Figure 1 and Figure 4 As shown, the unified management platform layer integrates a panoramic monitoring module, an automated operation and maintenance module, and a visualization interaction module. The panoramic monitoring module displays the CPU, memory, disk usage, and network bandwidth of each node in real time, and automatically triggers tiered alarms when equipment operating parameters are abnormal. The automated operation and maintenance module supports scripted task configuration to automate operations such as regular database backups and equipment status inspections. The visualization interaction module builds a 3D visualization interface based on UnityHDRP pipelines, displaying equipment health heatmaps, pipeline corrosion profiles, and AR inspection navigation annotations. It supports gesture-based virtual dashboard adjustments for simulation parameters. The edge collaboration layer deploys an industrial-grade edge computing gateway to achieve local preprocessing and real-time analysis of power plant field data, uploading only key fusion results and abnormal data to the cloud, reducing data transmission pressure and latency. The application output layer includes an equipment operation and maintenance module, a power generation optimization module, and a scheduling decision module. The equipment operation and maintenance module outputs predictive maintenance work orders, the power generation optimization module optimizes unit operating parameters based on fused data, and the scheduling decision module provides accurate data support for power grid scheduling.

[0035] Edge computing gateways perform local preprocessing and real-time analysis of field data, uploading only key fusion results and abnormal data. This reduces data transmission volume and controls latency to the millisecond level. The unified management platform enables panoramic monitoring (real-time display of resource usage and equipment parameters), automated operation and maintenance (scripted configuration backup and inspection), and visual interaction (3D interface + AR navigation), improving operation and maintenance efficiency. Edge and cloud collaboration ensures real-time on-site response and global unified management, adapting to distributed deployment scenarios in power plants.

[0036] In its daily operation, this invention first performs data acquisition and preprocessing, which involves collecting multi-source data using sensors, and then converting, cleaning, aligning, and standardizing the data to generate a high-quality dataset. Next, it performs layered data fusion and analysis, mining data value through three-level fusion and combining technologies such as digital twins and LSTM to complete equipment status simulation, lifespan prediction, and risk assessment. Finally, it performs edge and cloud collaboration, where edge gateways process data locally while the cloud provides unified monitoring and maintenance, achieving efficient collaboration. Furthermore, this process includes security protection and application, ensuring data and system security through end-to-end security measures, and the application layer outputs operation, maintenance, power generation, and dispatch decision-making results to support optimized power plant operation.

[0037] Example 2 This embodiment provides a power plant data fusion processing method based on a hyper-converged architecture, applied to the system described in Embodiment 1, and includes the following steps: Resource base deployment: Start the hyperconverged infrastructure layer, build a virtual resource pool through the virtualization engine, store data with the distributed storage module according to the replication mechanism and EC erasure coding, intelligently schedule hot and cold data to the corresponding storage media, and complete the construction of communication links at each level through software-defined networking.

[0038] Multi-source data acquisition: The multi-source data acquisition layer collects vibration, temperature, pressure, and infrared imaging data of power plant equipment and transmission lines in real time through a sensor cluster. The data is then standardized and accessed through a protocol conversion module that unifies heterogeneous protocols such as Modbus / TCP and MQTT.

[0039] Data preprocessing: The data preprocessing layer uses generative adversarial networks to correct sensor drift data, and trend extrapolation to complete missing data and remove noise; based on the IEEE1588v2 protocol and Kalman filtering, microsecond-level spatiotemporal alignment of multi-source data is achieved, and data units and formats are unified.

[0040] Layered data fusion: The layered data fusion layer sequentially completes the weighted average fusion of the data layer, the random forest and principal component analysis screening and dimensionality reduction of the feature layer, and the DS evidence theory and Bayesian network decision integration of the decision layer, outputting a highly reliable fusion result.

[0041] Intelligent Analysis and Assessment: The intelligent analysis layer constructs a digital twin model through multi-physics coupling, uses an LSTM network and a physical degradation model to predict the remaining service life of the equipment, and combines a risk matrix to complete dynamic risk assessment.

[0042] Edge collaboration and cloud management: The edge collaboration layer completes local preprocessing and real-time analysis of field data through industrial gateways, and only uploads key data to the cloud; the unified management platform layer realizes panoramic monitoring, automated operation and maintenance and 3D visualization interaction.

[0043] Security protection and application output: The security protection layer completes access control, full-link encryption and operation auditing according to the IEC62443 standard; the application output layer generates predictive maintenance work orders, optimizes unit operating parameters and provides data support for grid dispatch.

[0044] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the scope of the invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0045] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can be appropriately combined to form other embodiments that can be understood by those skilled in the art. The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

Claims

1. A power plant data fusion processing system based on a hyper-converged architecture, characterized in that, It includes a hyperconverged infrastructure layer, a multi-source data acquisition layer, a data preprocessing layer, a hierarchical data fusion layer, an intelligent analysis layer, a security protection layer, a unified management platform layer, an edge collaboration layer, and an application output layer. Each layer achieves bidirectional data interaction and resource collaboration through a software-defined network. The hyperconverged infrastructure layer provides integrated computing power, storage, and network resource support for the entire system.

2. The power plant data fusion processing system based on hyper-converged architecture according to claim 1, characterized in that, The hyperconverged infrastructure layer adopts a distributed architecture, integrating a virtualization engine, a distributed storage module, and a software-defined network controller. The virtualization engine supports KVM or VMware vSphere architecture, abstracting physical server resources into a virtual resource pool to achieve dynamic allocation and elastic scaling of CPU and memory resources. The distributed storage module uses a combination of replication mechanism and EC erasure coding to store power plant data, sharding the data and storing it on multiple nodes. It also configures a cold and hot data intelligent scheduling strategy, deploying high-frequency access data on SSD storage media and low-frequency data on HDD storage media.

3. The power plant data fusion processing system based on hyper-converged architecture according to claim 1, characterized in that, The multi-source data acquisition layer includes a sensor cluster, a protocol conversion module, and a data acquisition terminal. The sensor cluster covers the power plant system and transmission lines, including vibration sensors, temperature sensors, pressure sensors, and infrared array detectors. The protocol conversion module has a built-in OPCUA adapter to achieve unified conversion between different protocols such as Modbus / TCP and MQTT, and is compatible with data interaction between legacy PLC controllers and new intelligent devices.

4. The power plant data fusion processing system based on hyper-converged architecture according to claim 1, characterized in that, The data preprocessing layer includes a data cleaning module, a spatiotemporal alignment module, and a standardization module. The data cleaning module uses a generative adversarial network to correct sensor drift data, completes missing data through trend extrapolation, and removes invalid noise data. The spatiotemporal alignment module is based on the IEEE 1588v2 precision clock protocol and combines Kalman filtering to compensate for transmission delay, achieving microsecond-level synchronization between 10kHz sampling of vibration signals and 1Hz sampling of thermal parameters. The standardization module unifies the dimensions and formats of multi-source data and generates a standardized dataset that meets the requirements of fusion processing.

5. The power plant data fusion processing system based on a hyper-converged architecture according to claim 1, characterized in that, The hierarchical data fusion layer includes a data layer fusion module, a feature layer fusion module, and a decision layer fusion module; the data layer fusion module uses a weighted average algorithm to fuse multiple sets of data from the same source for the same monitoring object, preserving the details of the original data; The feature layer fusion module extracts time-domain features, frequency-domain features, and time-frequency features from the data, filters key degradation indicators through the random forest algorithm, and then simplifies the data dimensions through the principal component analysis algorithm. The decision layer fusion module integrates the preliminary decision results from various data sources based on DS evidence theory and Bayesian networks to achieve multi-dimensional verification of fault diagnosis and risk assessment.

6. The power plant data fusion processing system based on a hyper-converged architecture according to claim 1, characterized in that, The intelligent analysis layer includes a digital twin modeling module, a remaining useful life prediction module, and a dynamic risk assessment module. The digital twin modeling module uses multiphysics coupling simulation technology to construct thermal, mechanical, and control models to achieve high-fidelity simulation of the operating status of power plant equipment. The RUL prediction module adopts a hybrid framework that integrates LSTM networks and physical degradation models. The dynamic risk assessment module constructs a risk matrix to quantify the probability of failure and the severity of its consequences, guiding the prioritization of maintenance.

7. The power plant data fusion processing system based on hyper-converged architecture according to claim 1, characterized in that, The security protection layer conforms to the IEC62443 standard and includes an access control module, a data encryption module, and an operation auditing module. The access control module adopts a role-based access control mechanism to restrict the system operation scope of different maintenance personnel. The data encryption module uses SSL / TLS encryption for transmitted data and AES-256 encryption for stored data. It also uses snapshot and continuous data protection technologies to achieve real-time data backup and fault rollback. The operation auditing module records all system operation behaviors, generates traceable audit logs, and automatically triggers alarms for abnormal operations.

8. The power plant data fusion processing system based on hyper-converged architecture according to claim 1, characterized in that, The unified management platform integrates a panoramic monitoring module, an automated operation and maintenance module, and a visualization interaction module. The panoramic monitoring module displays the CPU, memory, disk usage, and network bandwidth of each node in real time, and automatically triggers tiered alarms when equipment operating parameters are abnormal. The automated operation and maintenance module supports scripted task configuration, enabling regular database backups and automated equipment status inspections. The visualization interaction module uses UnityHDRP pipeline to build a 3D visualization interface, displaying equipment health heatmaps, pipeline corrosion profiles, and AR inspection navigation annotations, and supports gesture-based virtual dashboard adjustments for simulation parameters.

9. The power plant data fusion processing system based on hyper-converged architecture according to claim 1, characterized in that, The edge collaboration layer deploys an industrial-grade edge computing gateway to enable local preprocessing and real-time analysis of power plant field data, uploading only key fusion results and abnormal data to the cloud, thus reducing data transmission pressure and latency. The application output layer includes an equipment operation and maintenance module, a power generation optimization module, and a scheduling decision module. The equipment operation and maintenance module outputs predictive maintenance work orders, the power generation optimization module optimizes unit operating parameters based on fused data, and the scheduling decision module provides accurate data support for power grid scheduling.

10. A power plant data fusion processing method based on a hyper-converged architecture, characterized in that, This method is based on the power plant data fusion processing system based on a hyper-converged architecture as described in any one of claims 1 to 9, comprising: The hyperconverged infrastructure layer first integrates and dynamically allocates computing power, storage, and network resources; the multi-source data acquisition layer collects multi-dimensional monitoring data from power plant equipment and transmission lines and completes unified protocol conversion; the data preprocessing layer cleans, aligns, and standardizes the collected data to generate a standardized dataset; the layered data fusion layer sequentially performs three-level fusion processing of the data layer, feature layer, and decision layer to extract the core value of the data; the intelligent analysis layer completes digital twin modeling, remaining service life prediction, and dynamic risk assessment based on the fused data; the security protection layer implements access control, data encryption, and operation auditing throughout the process to ensure system and data security; the unified management platform layer realizes panoramic system monitoring, automated operation and maintenance, and three-dimensional visualization interactive management; the edge collaboration layer completes local preprocessing and real-time analysis of on-site data, uploading only key data to the cloud; and the application output layer outputs relevant instructions and data for equipment operation and maintenance, power generation optimization, and scheduling decisions based on the processing and analysis results to support the efficient and stable operation of the power plant.

Images