Water-turbine generator set digital twinning system for state monitoring and diagnosis

Abstract

The invention discloses a water-turbine generator set digital twinning system for state monitoring and diagnosis, which comprises a water-turbine generator entity layer, the water-turbine generator entity layer is in communication connection with a sensor system, the sensor system is in communication connection with a digital twinning model layer, and the digital twinning model layer is in communication connection with a twinning data processing layer. The twinborn data processing layer is in communication connection with the digital twinborn decision-making layer; the hydro-generator entity layer, the sensor system, the digital twinborn model layer, the twinborn data processing layer and the digital twinborn decision-making layer are all in bidirectional communication connection with the hydro-generator set intelligent operation and maintenance platform and are in communication connection with the cloud computing cloud service module. The system is used for solving the problems that in the prior art, a control strategy cannot clearly reflect the corresponding relation between faults and symptoms, system data is too single, accuracy is insufficient, and efficiency is poor. The fault of the water-turbine generator set can be efficiently, accurately and quickly judged and predicted in the real-time operation process of the water-turbine generator set.

Description

Technical Field

[0001] The present invention belongs to the technical field of status and fault warning of a hydro-generator set, and in particular relates to a digital twin system of a hydro-generator set for status monitoring and diagnosis. Background Art

[0002] With the rapid development of my country's hydropower industry, the installed capacity and single-unit capacity of hydro-turbine generators are getting larger and larger, and their power generation accounts for an increasingly higher proportion in the power system. The stable operation of hydro-turbine generator sets is becoming increasingly important to the safety and economic operation of the power grid. However, for a long time, hydro-turbine generator sets still use a maintenance system of regular minor, medium and major repairs according to a plan. Such planned maintenance has defects such as over-maintenance and insufficient maintenance. With the emergence of a large number of giant hydro-turbine generator sets, the drawbacks of the current maintenance system are gradually emerging. In order to improve the pertinence of maintenance and ensure the safe, economical, long-term and stable operation of hydropower stations, the research and development of excellent hydropower unit status monitoring and fault diagnosis systems is imminent. This is not only the direction of the development of intelligent operation of hydropower stations, but also has extremely important significance for the power industry and even the entire national economy. At present, the systems for monitoring the status of hydro-turbine generator sets have the following problems: most control strategies cannot clearly point out the correspondence between faults and symptoms, and cannot intuitively see the system simulation models under normal and fault conditions; the system database is relatively single, the method for judging system faults is not perfect, and the results are not accurate enough; cloud computing services are not used to provide computing power support for the system, resulting in low computing power of the system and high project investment costs, and it is not convenient for staff to remotely monitor the operating status of the system in different regions; in the process of judging faults, the system will generate a large amount of data, and traditional detection methods will consume a lot of computing resources when processing these massive data, lacking an efficient agent model; therefore, it is necessary to design a digital twin system, device, electronic equipment and storage medium for hydro-turbine generator sets for status monitoring and diagnosis to solve the above problems. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a digital twin system of a hydro-turbine generator set for condition monitoring and diagnosis, which is used to solve the problems in the prior art that the control strategy cannot clearly reflect the correspondence between faults and symptoms, the system data is too single, the accuracy is insufficient, and the efficiency is poor; it can efficiently, accurately and quickly judge and predict the occurrence of faults in the hydro-turbine generator set during the real-time operation of the hydro-turbine generator set.

[0004] In order to achieve the above technical effects, the technical solution adopted by the present invention is: A digital twin system for a hydro-turbine generator set for condition monitoring and diagnosis includes a hydro-turbine generator physical layer, the hydro-turbine generator physical layer is communicatively connected to a sensor system, the sensor system is communicatively connected to a digital twin model layer, the digital twin model layer is communicatively connected to a twin data processing layer, and the twin data processing layer is communicatively connected to a digital twin decision layer; the hydro-turbine generator physical layer, the sensor system, the digital twin model layer, the twin data processing layer, and the digital twin decision layer are all bidirectionally communicatively connected to an intelligent operation and maintenance platform of the hydro-turbine generator set to form a digital twin system, and the digital twin system is communicatively connected to a cloud computing cloud service module.

[0005] Preferably, the physical layer of the hydro-generator includes an excitation system, a hydraulic system, an energy system, a control system, an auxiliary system, an electrical system, a propulsion system, a power generation system, a rotating shaft, a volute and blades.

[0006] Preferably, the sensor system includes a data acquisition module, a signal processing module, a control module and a communication module distributed on the physical layer of the hydro-turbine generator; the sensor system collects real-time operating condition data and real-time status monitoring data of key components of the hydro-turbine generator, including status data of vibration, deformation, temperature, magnetic flux, pressure, speed, current and power, and uploads these data in real time to the intelligent operation and maintenance platform and digital twin model layer of the hydro-turbine generator set, and transmits the system control signal to the sensor system through the intelligent operation and maintenance platform of the hydro-turbine generator set to control it.

[0007] Preferably, the digital twin model layer collects various data from the sensor system through a device interaction interface system that adopts the OPC UA protocol, and uses these data as boundary conditions for system simulation to establish a visual digital twin model; the digital twin model layer receives the control signal of the hydropower generator set of the intelligent operation and maintenance platform and generates the required model, and uploads the generated model and comparison results to the intelligent operation and maintenance platform.

[0008] Furthermore, the OPC UA protocol has good system compatibility, high security and wide industry recognition, and can ensure efficient and secure communication between various devices and systems.

[0009] Preferably, the visualized digital twin model includes finite element models and mechanism models of the system under normal and faulty conditions, as well as a lightweight proxy model constructed using a machine learning algorithm in combination with real-time sensor information, to realize the functions of processing massive data, responding to real-time changes, and reducing computing resource consumption.

[0010] Preferably, the lightweight proxy model is constructed by: Build a high-fidelity simulation model of a hydro-generator set; Choose the appropriate proxy model type, such as a neural network, based on the actual data set; Construct an agent model based on multi-source fusion perception; The proxy model is applied to actual detection.

[0011] Preferably, the twin data processing layer establishes multi-physics field simulation and boundary models through high-performance computing power, including electromagnetic-thermal coupling models, thermal-structural coupling models, and structural-fluid coupling models; The twin data processing layer compares the real-time model and theoretical model in the database, optimizes and adjusts the existing model, and realizes real-time, two-way interaction between the virtual model and the system entity layer through the virtual interaction and collaboration mechanism and the intelligent operation and maintenance platform of the hydro-generator set. The correctness of the system simulation is verified through the real-time collected system operation data, and the boundary conditions and algorithm parameters are optimized to adjust the simulation model. The twin data processing layer intervenes or adjusts the hydro-generator system entities based on the early warning information from the virtual model.

[0012] Preferably, the virtual interaction and collaboration mechanism is used to realize real-time, two-way interaction between the virtual model and the system entity layer. The virtual generator set model collects physical data from the system entity in real time to verify whether the simulation model is accurate, and adjusts the simulation model by optimizing boundary conditions and algorithm parameters based on this goal. At the same time, the hydropower generator set system entity can be intervened or adjusted based on the early warning information from the virtual model.

[0013] Preferably, the digital twin decision layer is used to monitor the interaction and overall performance of all models, including monitoring and analyzing the real-time operating status of the physical layer of the hydro-generator and the simulation model, diagnosing the system faults by using the fault detection algorithm and intelligent diagnosis algorithm in the algorithm library combined with expert experience and intelligent analysis and evaluation methods, uploading the diagnosis results to the intelligent operation and maintenance platform of the hydro-generator set, and notifying the operation and maintenance personnel through the diagnosis results to physically troubleshoot and repair the system faults.

[0014] Preferably, the intelligent operation and maintenance platform of the hydro-generator set is used to realize data interaction and regulation of various parts of the system, integrate data from different sources to form a unified and coordinated view, and simplify data access and reduce the complexity of interaction between systems; specifically, it includes: Receive and store data transmitted by various parts of the system in real time, including sensor data, established simulation models and mechanism models, and fault monitoring data; Combined with historical performance data, the data is processed through intelligent analysis and evaluation methods, and then the information is classified and sent to various parts of the system, including decision instructions from the digital twin decision layer, early warning information from the twin data processing layer, and data collected by the sensor system; Intelligent algorithms are used to optimize decisions to determine existing operating conditions and operating monitoring data, and to clarify the scope and characteristics of additional operating conditions and monitoring data that need to be added.

[0015] Furthermore, cloud computing and cloud services do not require huge investments in building supercomputing centers, but can provide the large amount of computing power required by the system through cloud computing; the use of cloud services can serve users in different regions and allow staff to remotely monitor the operation of the system in real time.

[0016] The beneficial effects of the present invention are as follows: 1. This application establishes a visual finite element model and mechanism model of normal and faulty data under multi-physical field coupling; the constructed lightweight proxy model can process massive data, respond to real-time changes, and reduce computing resource consumption. It can be compared with a real-time updated database and use a rich intelligent algorithm to more finely distinguish the relationship between faults and symptoms; 2. This application receives and integrates different data through an intelligent operation and maintenance platform, which can establish a more sophisticated and rich database and simplify the complexity of data interaction, so that the digital twin system can give more accurate causes of failures and solutions.

[0017] 3. This application adopts cloud computing and cloud service methods to solve the problems of low computing power and high cost of building supercomputing centers in traditional methods. In addition, the use of cloud service methods allows users in different regions to easily observe the system operation status. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a schematic diagram of the system architecture of this application; Figure 2 It is a schematic diagram of the structure of the physical layer of the hydro-generator of the present application; Figure 3 is a schematic diagram of the architecture of the sensor system of the present application; Figure 4 It is a schematic diagram of the architecture of the digital twin model layer of this application; Figure 5 It is a schematic diagram of the architecture of the twin data processing layer of this application; Figure 6 It is a schematic diagram of the architecture of the digital twin decision layer of this application; Figure 7 It is a schematic diagram of the architecture of the intelligent operation and maintenance platform of the hydro-generator set of the present application; Figure 8 It is a system architecture diagram of an example of a digital twin system for a hydro-turbine generator set for condition monitoring and diagnosis in an embodiment of the present application. DETAILED DESCRIPTION

[0019] Embodiment 1: like Figure 1As shown, a digital twin system of a hydro-turbine generator set for condition monitoring and diagnosis includes a hydro-turbine generator physical layer, the hydro-turbine generator physical layer is communicatively connected to the sensor system, the sensor system is communicatively connected to the digital twin model layer, the digital twin model layer is communicatively connected to the twin data processing layer, and the twin data processing layer is communicatively connected to the digital twin decision layer; the hydro-turbine generator physical layer, the sensor system, the digital twin model layer, the twin data processing layer and the digital twin decision layer are all bidirectionally communicatively connected to the intelligent operation and maintenance platform of the hydro-turbine generator set to form a digital twin system, and the digital twin system is communicatively connected to the cloud computing cloud service module.

[0020] The physical layer of the hydro-generator includes the excitation system, hydraulic system, energy system, control system, auxiliary system, electrical system, propulsion system, power generation system, as well as the shaft, volute and blades.

[0021] Preferably, the sensor system includes a data acquisition module, a signal processing module, a control module and a communication module distributed on the physical layer of the hydro-turbine generator; the sensor system collects real-time operating condition data and real-time status monitoring data of key components of the hydro-turbine generator, including status data of vibration, deformation, temperature, magnetic flux, pressure, speed, current and power, and uploads these data in real time to the intelligent operation and maintenance platform and digital twin model layer of the hydro-turbine generator set, and transmits the system control signal to the sensor system through the intelligent operation and maintenance platform of the hydro-turbine generator set to control it.

[0022] Preferably, the digital twin model layer collects various data from the sensor system through a device interaction interface system that adopts the OPC UA protocol, and uses these data as boundary conditions for system simulation to establish a visual digital twin model; the digital twin model layer receives the control signal of the hydropower generator set of the intelligent operation and maintenance platform and generates the required model, and uploads the generated model and comparison results to the intelligent operation and maintenance platform.

[0023] Furthermore, the OPC UA protocol has good system compatibility, high security and wide industry recognition, and can ensure efficient and secure communication between various devices and systems.

[0024] Preferably, the visualized digital twin model includes finite element models and mechanism models of the system under normal and faulty conditions, as well as a lightweight proxy model constructed using a machine learning algorithm in combination with real-time sensor information, to realize the functions of processing massive data, responding to real-time changes, and reducing computing resource consumption.

[0025] Preferably, the lightweight proxy model is constructed by: Build a high-fidelity simulation model of a hydro-generator set; Choose the appropriate proxy model type, such as a neural network, based on the actual data set; Construct an agent model based on multi-source fusion perception; The proxy model is applied to actual detection.

[0026] Preferably, the twin data processing layer establishes multi-physics field simulation and boundary models through high-performance computing power, including electromagnetic-thermal coupling models, thermal-structural coupling models, and structural-fluid coupling models; The twin data processing layer compares the real-time model and theoretical model in the database, optimizes and adjusts the existing model, and realizes real-time, two-way interaction between the virtual model and the system entity layer through the virtual interaction and collaboration mechanism and the intelligent operation and maintenance platform of the hydro-generator set. The correctness of the system simulation is verified through the real-time collected system operation data, and the boundary conditions and algorithm parameters are optimized to adjust the simulation model. The twin data processing layer intervenes or adjusts the hydro-generator system entities based on the early warning information from the virtual model.

[0027] Preferably, the virtual interaction and collaboration mechanism is used to realize real-time, two-way interaction between the virtual model and the system entity layer. The virtual generator set model collects physical data from the system entity in real time to verify whether the simulation model is accurate, and adjusts the simulation model by optimizing boundary conditions and algorithm parameters based on this goal. At the same time, the hydropower generator set system entity can be intervened or adjusted based on the early warning information from the virtual model.

[0028] Preferably, the digital twin decision layer is used to monitor the interaction and overall performance of all models, including monitoring and analyzing the real-time operating status of the physical layer of the hydro-generator and the simulation model, diagnosing the system faults by using the fault detection algorithm and intelligent diagnosis algorithm in the algorithm library combined with expert experience and intelligent analysis and evaluation methods, uploading the diagnosis results to the intelligent operation and maintenance platform of the hydro-generator set, and notifying the operation and maintenance personnel through the diagnosis results to physically troubleshoot and repair the system faults.

[0029] Preferably, the intelligent operation and maintenance platform of the hydro-generator set is used to realize data interaction and regulation of various parts of the system, integrate data from different sources to form a unified and coordinated view, and simplify data access and reduce the complexity of interaction between systems; specifically, it includes: Receive and store data transmitted by various parts of the system in real time, including sensor data, established simulation models and mechanism models, and fault monitoring data; Combined with historical performance data, the data is processed through intelligent analysis and evaluation methods, and then the information is classified and sent to various parts of the system, including decision instructions from the digital twin decision layer, early warning information from the twin data processing layer, and data collected by the sensor system; Intelligent algorithms are used to optimize decisions to determine existing operating conditions and operating monitoring data, and to clarify the scope and characteristics of additional operating conditions and monitoring data that need to be added.

[0030] Furthermore, cloud computing and cloud services do not require huge investments in building supercomputing centers, but can provide the large amount of computing power required by the system through cloud computing; the use of cloud services can serve users in different regions and allow staff to remotely monitor the operation of the system in real time.

[0031] Embodiment 2: The physical layer of the turbine generator set includes the excitation system, hydraulic system, energy system, control system, auxiliary system, electrical system, propulsion system, power generation system, as well as the shaft, volute, blades, etc. Figure 2 shown.

[0032] The digital twin system is used to detect real-time faults in the hydro-turbine generator set, including turbine faults, generator electrical system faults, and shaft system faults.

[0033] The real-time operating condition data and real-time status monitoring data of the key components of the hydro-generator are collected through the data acquisition module, signal processing module, control module and communication module distributed on the physical layer of the hydro-generator, including vibration noise, deformation, temperature, magnetic flux, pressure, speed, voltage, current, power, flow, displacement and gas in oil. Using multiple monitoring methods, the dynamic data monitored are statistically summarized to specifically characterize the evolution of faults, and then transmitted to the digital twin space, that is, its next digital twin model layer and the intelligent operation and maintenance platform of the hydro-generator set responsible for the overall control system; sensor systems such as Figure 3 shown.

[0034] The digital twin model refers to a virtual model that characterizes the normal and faulty behaviors of a system. In terms of status monitoring and diagnosis, the abnormal fault status of system components has become the focus of attention. The fault behavior model is a digital projection of the system fault behavior in virtual space, which can establish a mapping relationship between the state and fault under the fault behavior. Virtual simulation based on the fault behavior model can generate fault data under various fault modes, providing more complete data support for the diagnosis of the status monitoring of the hydropower system.

[0035] Digital twin model layers such as Figure 4As shown. The data interaction interface uses OPCUA as the main data interaction protocol to ensure efficient, real-time, safe and reliable data interaction. The data monitored by the sensor is uploaded to the digital twin model layer efficiently and accurately. The intelligent operation and maintenance platform of the hydro-turbine generator set performs regulation and uses the data collected by the sensor as the boundary conditions of the system simulation to establish a visual multi-physics coupling simulation model, including finite element models and mechanism models under normal and faulty systems; based on the constructed high-fidelity physical model, combined with real-time sensor information, a high-fidelity lightweight proxy model is constructed through machine learning algorithms, which can realize the functions of processing massive data, responding to real-time changes, and reducing computing resource consumption. The finite element model includes: electromagnetic model, thermal model, structural model and fluid dynamic model. In addition, the digital twin model layer needs to interact with the intelligent operation and maintenance platform, receive its control signals and generate the required models, and transmit the generated models to the twin data processing layer or upload them to the intelligent operation and maintenance platform.

[0036] The twin data layer is a bridge between the physical layer of the hydro-turbine generator system and the digital twin model layer, covering the physical data of the system entity, simulation data (consisting of normal and fault simulation data), digital twin application data, and the fusion and derivative data between them. The bridging method between the system entity and the digital twin model is implemented according to the virtual interaction and collaboration mechanism. The main function is to realize real-time, two-way interaction between the virtual model and the system entity layer; the twin data processing layer is as follows: Figure 5 shown.

[0037] Based on the virtual-real interaction and collaboration mechanism of data fusion, the virtual model is a true expression of the system entity, so the digital twin can grasp the operating status of the system. Moreover, under the hierarchical digital twin architecture, there is a point-to-point correspondence between the virtual model and the system entity. Therefore, the digital twin can not only grasp the sensor signals perceived by the outside world, but also gain insight into the information of the internal components of the system to achieve more comprehensive operating status monitoring. The virtual generator set model collects physical data from the system entity in real time to verify whether the simulation model is accurate, and adjusts the boundary conditions and algorithm parameters to adjust the simulation model based on this goal. At the same time, it can intervene or adjust the hydropower generator set system entity based on the early warning information from the virtual model.

[0038] Monitor the interaction and overall performance of all models, monitor and analyze the real-time operating status of the physical layer of the hydro-generator and the simulation model, diagnose the system faults by using the fault detection algorithm and intelligent diagnosis algorithm in the algorithm library combined with expert experience and intelligent analysis and evaluation methods, and give relevant instructions for decision-making execution, including active protection of the unit protection device and troubleshooting of operators and maintenance personnel, and upload the diagnosis results to the intelligent operation and maintenance platform of the hydro-generator unit. The operation and maintenance personnel can use the diagnosis results to physically troubleshoot and repair the system faults. The accuracy of information transmission from the generation of the fault to the data twin layer and the accuracy of the analysis and evaluation data will be transmitted to the management department to further correct and improve the digital twin. See the digital twin decision layer model. Figure 6 .

[0039] Realize data interaction and regulation of various parts of the system, integrate data from different sources to form a unified and coordinated view, simplify data access, and reduce the complexity of interaction between systems. Specifically, it includes real-time reception and storage of data transmitted by various parts of the system, including sensor data, established simulation models and mechanism models, and fault monitoring data, etc., and combines historical performance data to process the data through intelligent analysis and evaluation methods, and then classify the information and send it to various parts of the system, including decision instructions of the digital twin decision layer, early warning information of the twin data processing layer, and data collected by the sensor system; use intelligent algorithms to optimize decisions to determine existing operating conditions and operating monitoring data, and clarify the scope and characteristics of the conditions and monitoring data that need to be added. See the intelligent operation and maintenance platform for hydropower generator sets. Figure 7 shown.

[0040] In this digital twin system of hydroelectric generator sets, the traditional computing center is not used to provide computing power support for the system, but cloud computing is used to provide computing power support for it, which saves the huge investment required to establish a computing center and provides deeper computing power; in addition, users can flexibly increase or decrease computing resources such as processing power, storage space and bandwidth according to their needs to meet different business needs. Cloud services usually have a wide coverage, allowing system staff to monitor the operation status of the system in real time in different places.

[0041] Embodiment three: According to the scheme of the present invention, a digital twin system example of a hydro-turbine generator set for condition monitoring and diagnosis is constructed, such as Figure 8 shown.

[0042] Step 1: Use multiple monitoring methods to collect real-time operating condition data and real-time status monitoring data of key components of the hydro-generator through data acquisition modules, signal processing modules, control modules and communication modules distributed on the physical layer of the hydro-generator, including vibration, deformation, temperature, magnetic flux, pressure, speed, current, power and other status data, and upload these data to the intelligent operation and maintenance platform and digital twin model layer of the hydro-generator set in real time, and transmit the system control signal to the sensor system through the intelligent operation and maintenance platform of the hydro-generator set to control it.

[0043] Step 2: Based on cloud computing services, the OPCUA protocol data interaction interface is used to efficiently and accurately upload the data monitored by the sensors to the digital twin model layer. The intelligent operation and maintenance platform of the hydro-turbine generator set is regulated and the data collected by the sensors is used as the boundary conditions of the system simulation to establish a visual multi-physics coupling simulation model, including finite element models and mechanism models under normal and faulty systems, and compared with the finite element models and mechanism models in the system database. Secondly, based on the constructed high-fidelity physical model, a high-fidelity lightweight proxy model is constructed through a machine learning algorithm in combination with real-time sensor information. In addition, the digital twin model layer receives the control signal of the hydro-turbine generator set from the intelligent operation and maintenance platform and generates the required model, and uploads the generated model and comparison results to the intelligent operation and maintenance platform.

[0044] Step 3: Based on cloud computing services, the twin data processing layer realizes real-time, two-way interaction between the virtual model and the system entity layer. Through high-performance computing power, multi-physics field simulation and boundary models including electromagnetic-thermal coupling model, thermal-structural coupling model and structural-fluid coupling model are established, and the real-time model and theoretical model in the database are compared to optimize and adjust the existing model. Through virtual interaction and collaboration mechanism, the intelligent operation and maintenance platform of the hydro-turbine generator set realizes real-time, two-way interaction between the virtual model and the system entity layer. The correctness of the system simulation is verified through the real-time collected system operation data, and the boundary conditions and algorithm parameters are optimized to adjust the simulation model. At the same time, the hydro-turbine generator set system entity can be intervened or adjusted based on the early warning information from the virtual model.

[0045] Step 4: Based on cloud computing services, the digital twin decision layer monitors the interaction and overall performance of all models. The virtual generator set model collects physical data from system entities in real time to verify whether the simulation model is accurate, and adjusts the simulation model by optimizing boundary conditions and algorithm parameters. At the same time, the hydropower generator set system entity can be intervened or adjusted based on the early warning information from the virtual model.

[0046] Step 5: Based on cloud computing services, the intelligent operation and maintenance platform of the hydro-generator sets realizes data interaction and regulation of various parts of the system, integrates data from different sources to form a unified and coordinated view, simplifies data access, and reduces the complexity of interaction between systems. Specifically, it includes real-time reception and storage of data transmitted by various parts of the system, including sensor data, established simulation models and mechanism models, and fault monitoring data, etc., combined with historical performance data, the data is processed through intelligent analysis and evaluation methods, and then the information is classified and sent to various parts of the system, including decision instructions of the digital twin decision layer, early warning information of the twin data processing layer, data collected by the sensor system, etc.; intelligent algorithms are used to optimize decisions to determine existing operating conditions and operating monitoring data, and to clarify the scope and characteristics of the conditions and monitoring data that need to be added.

Claims

1. A digital twin system for hydro-turbine generator sets for condition monitoring and diagnosis, characterized in that: It includes a hydro-turbine generator physical layer, the hydro-turbine generator physical layer is communicated with the sensor system, the sensor system is communicated with the digital twin model layer, the digital twin model layer is communicated with the twin data processing layer, and the twin data processing layer is communicated with the digital twin decision layer; the hydro-turbine generator physical layer, sensor system, digital twin model layer, twin data processing layer and digital twin decision layer are all communicated with the hydro-turbine generator set intelligent operation and maintenance platform in a two-way manner to form a digital twin system, and the digital twin system is communicated with the cloud computing cloud service module.

2. A digital twin system for condition monitoring and diagnosis of a hydro-generator set according to claim 1, characterized in that: The physical layer of the hydro-generator includes the excitation system, hydraulic system, energy system, control system, auxiliary system, electrical system, propulsion system, power generation system, as well as the shaft, volute and blades.

3. The digital twin system for condition monitoring and diagnosis of a hydro-generator set according to claim 1 is characterized in that: The sensor system includes a data acquisition module, a signal processing module, a control module and a communication module distributed on the physical layer of the hydro-generator; the sensor system collects real-time operating condition data and real-time status monitoring data of key components of the hydro-generator, including vibration, deformation, temperature, magnetic flux, pressure, speed, current and power status data, and uploads these data in real time to the intelligent operation and maintenance platform and digital twin model layer of the hydro-generator set, and transmits the system control signal to the sensor system through the intelligent operation and maintenance platform of the hydro-generator set for control.

4. The digital twin system for condition monitoring and diagnosis of a hydro-generator set according to claim 1 is characterized in that: The digital twin model layer collects various data from the sensor system through the device interaction interface system using the OPCUA protocol, and uses these data as the boundary conditions of the system simulation to establish a visual digital twin model; The digital twin model layer receives the control signal of the hydro-generator set from the intelligent operation and maintenance platform and generates the required model, and uploads the generated model and comparison results to the intelligent operation and maintenance platform.

5. A digital twin system for condition monitoring and diagnosis of a hydro-generator set according to claim 4, characterized in that: The visualized digital twin model includes finite element models and mechanism models under normal and faulty system conditions, as well as a lightweight proxy model built using machine learning algorithms in combination with real-time sensor information. It is used to process massive amounts of data, respond to real-time changes, and reduce computing resource consumption.

6. A digital twin system for condition monitoring and diagnosis of a hydro-generator set according to claim 5, characterized in that: The construction method of the lightweight proxy model is: Build a high-fidelity simulation model of a hydro-generator set; Choose the appropriate proxy model type based on the actual data set; Construct an agent model based on multi-source fusion perception; The proxy model is applied to actual detection.

7. A digital twin system for condition monitoring and diagnosis of a hydro-generator set according to claim 1, characterized in that: The twin data processing layer uses high-performance computing power to establish multi-physics field simulation and boundary models, including electromagnetic-thermal coupling models, thermal-structural coupling models, and structural-fluid coupling models; The twin data processing layer compares the real-time model and theoretical model in the database, optimizes and adjusts the existing model, and realizes real-time, two-way interaction between the virtual model and the system entity layer through the virtual interaction and collaboration mechanism and the intelligent operation and maintenance platform of the hydro-generator set. The correctness of the system simulation is verified through the real-time collected system operation data, and the boundary conditions and algorithm parameters are optimized to adjust the simulation model. The twin data processing layer intervenes or adjusts the hydro-generator system entities based on the early warning information from the virtual model.

8. The digital twin system for condition monitoring and diagnosis of a hydro-generator set according to claim 7 is characterized in that: The virtual interaction and collaboration mechanism is used to realize real-time, two-way interaction between the virtual model and the system entity layer. The virtual generator set model collects physical data from the system entity in real time to verify whether the simulation model is accurate, and adjusts the simulation model by optimizing boundary conditions and algorithm parameters. At the same time, the hydropower generator set system entity can be intervened or adjusted based on the early warning information from the virtual model.

9. The digital twin system for condition monitoring and diagnosis of a hydro-generator set according to claim 1, characterized in that: The digital twin decision layer is used to monitor the interaction and overall performance of all models, including monitoring and analyzing the real-time operating status of the physical layer of the hydro-turbine generator and the simulation model. It uses the fault detection algorithm and intelligent diagnosis algorithm in the algorithm library combined with expert experience and intelligent analysis and evaluation methods to diagnose the system's faults, and uploads the diagnosis results to the intelligent operation and maintenance platform of the hydro-turbine generator set. The operation and maintenance personnel are notified through the diagnosis results to physically troubleshoot and repair the system's faults.

10. A digital twin system for condition monitoring and diagnosis of a hydro-generator set according to claim 1, characterized in that: The intelligent operation and maintenance platform of hydro-turbine generator sets is used to realize data interaction and regulation of various parts of the system, integrate data from different sources, and form a unified and coordinated view; specifically, it includes: Receive and store data transmitted by various parts of the system in real time, including sensor data, established simulation models and mechanism models, and fault monitoring data; Combine historical performance data with intelligent analysis and evaluation methods to process the data and then classify the information and send it to various parts of the system, including decision instructions from the digital twin decision layer, early warning information from the twin data processing layer, and data collected by the sensor system; Intelligent algorithms are used to optimize decisions to determine existing operating conditions and operating monitoring data, and to clarify the scope and characteristics of additional operating conditions and monitoring data that need to be added.

Images