Digital twin model for turbine component testing

Abstract

This invention belongs to the field of digital twin technology, specifically relating to a digital twin model for turbine component testing. The specific technical solution is as follows: the main model architecture comprises three layers connected sequentially: a physical information network layer, a physical information coupling network layer, and a mapping network layer. The output of the physical information network layer is connected to the physical information coupling network layer, and the output of the physical information coupling network layer is connected to the mapping network layer. This model is used for online operation and maintenance monitoring, fault diagnosis, and health maintenance and management of experimental equipment during the testing process of aero-engine turbine components. It provides early warning of potential damage to turbine hydraulic dynamometer equipment, enabling proactive maintenance, extending the service life of the dynamometer equipment, reducing maintenance costs, and ultimately improving the stability, safety, and economy of the turbine testing process.

Description

Technical Field

[0001] This invention belongs to the field of digital twin technology, specifically relating to a digital twin model for testing turbine components. Background Technology

[0002] Physical experiments on turbine components are essential for the development of new aero-engines and the modification / improvement design of existing engines. As the most direct power-generating component of an aero-engine, the turbine plays a crucial role in all aero-engine tests. To ensure the timely and high-quality completion of the entire engine development, it is necessary to guarantee and optimize the safety, stability, accuracy, and economy of the testing process for core components such as turbines.

[0003] Traditional physical testing of turbine components mainly involves measuring the turbine's key performance parameters and controlling the experimental conditions using dynamometers (devices that measure the output torque or drive torque of machinery). Common dynamometers include hydraulic, eddy current, and electric types. Among them, hydraulic dynamometers have advantages such as small size, large power reserve, and high accuracy, and are widely used in aero-engine testing and power rotating machinery testing.

[0004] A hydraulic dynamometer mainly consists of a force-measuring mechanism, a water supply system, and a brake. The force-measuring mechanism measures the engine's output power; the water supply system supplies and discharges different amounts of water according to different power requirements; and the brake absorbs the engine's output power. The hydraulic dynamometer utilizes the frictional resistance generated between water and a moving body to absorb engine power. Its basic components are as follows: Figure 1 As shown, the hydraulic dynamometer shaft and the turbine shaft are connected by coupling 2 to transfer power between the hydraulic dynamometer system and the turbine system. When the main shaft 4 drives the rotor 1 to rotate, the water entering the dynamometer will generate friction and rapid impact with the rotor, thus consuming the power of the turbine test piece. The mechanical energy output by the turbine test piece is converted into the heat energy of the water and carried away by the water. The power is absorbed by the viscous shear force in the boundary layer between the adjacent rotor 1 and the stator surface. The resulting resistance dampens the rotation and is converted into an equal force on the dynamometer housing 3. A torque sensor is arranged on the dynamometer housing 3. The power of the turbine test piece can be obtained by measuring the torque and shaft speed. In addition, the lubrication system is used to provide oil lubrication and cooling for the high-speed rotating bearings in both the dynamometer and the turbine body. The two systems are connected by connecting the main shaft of the hydraulic dynamometer and the main shaft of the turbine dynamometer through the coupling.

[0005] Currently, there are two main methods for performance tracking during turbine component testing.

[0006] The first approach involves experienced test engineers subjectively judging the operational status of the turbine testing system and turbine hydraulic dynamometer by observing test data and the test condition of the test specimens during the test. For example, test engineers can determine if there are problems during the test process by observing whether the monitored parameters change drastically or exceed the specified limits, and then take appropriate actions such as starting, stopping, or adjusting the equipment to ensure the accurate and safe operation of the experiment. Regarding equipment health management, because high-speed hydraulic dynamometers used in aero-engines are usually purchased from the UK and the US, and their internal systems are complex with unclear mechanisms and parameter relationships, the current practice is to incur significant costs by hiring the equipment supplier for maintenance.

[0007] The aforementioned method has the following technical problems: First, manual testing, monitoring, and tracking rely on experienced testing technicians, and training such technicians requires a long-term investment of time and resources. Second, this subjective, experience-based approach cannot provide accurate and effective results for complex problems. Furthermore, the health management of turbine hydraulic dynamometers currently faces challenges due to the fact that the design and manufacturing technology of these devices is controlled by foreign companies in the UK and the US, resulting in high maintenance risks, difficulties, and high costs.

[0008] The second approach is to establish a digital twin model to monitor the operating status of the dynamometer. Existing digital twin modeling techniques traditionally employ a purely data-driven approach without incorporating actual physical models. However, in real turbine component testing systems, the turbine thermodynamic system, rotor transmission system, hydrodynamic system, and lubrication system are intricately coupled. Therefore, if a traditional purely data-driven approach is used, the calculation results are prone to deviating from real-world physical laws. Furthermore, deep learning models, which rely on purely data-driven methods, do not incorporate engine physical models and architectures. Consequently, they lack the constraints of actual physical rules, leading to training results that easily deviate from physical rules, resulting in large errors and unreliable simulations by the digital twin.

[0009] Therefore, if a digital twin model based on the working principle of turbine hydraulic dynamometers can be provided, it will have excellent prospects for industrial applications. Summary of the Invention

[0010] To address the aforementioned technical issues, this invention proposes a digital twin model based on the working principle of a turbine hydraulic dynamometer. This model is used for online operation and maintenance monitoring, fault diagnosis, and health maintenance and management of experimental equipment during the testing process of aero-engine turbine components. It provides early warning of potential damage to the turbine hydraulic dynamometer, enabling proactive maintenance, extending the service life of the dynamometer, reducing maintenance costs, and ultimately improving the stability, safety, and economy of the turbine testing process.

[0011] To achieve the above-mentioned objectives, the technical solution adopted by the present invention is: a digital twin model for turbine component testing, the main body of the model architecture including three layers connected in sequence: a physical information network layer, a physical information coupling network layer, and a mapping network layer; the output end of the physical information network layer is connected to the physical information coupling network layer, and the output end of the physical information coupling network layer is connected to the mapping network layer.

[0012] The physical information network layer includes a time information network layer, an intake pipeline network layer, a turbine network layer, a front bearing network layer, a water system network layer, a rear bearing network layer, and a shaft power system network layer.

[0013] The physical information coupling network layer adopts a temporal network structure, which connects the physical information of the physical information network layer in a sequential manner in time and space, extracts the physical correlation and physical features between the test parameters of the turbine hydraulic dynamometer system, and stores them in the neural network.

[0014] The mapping network layer maps the information of the physical information coupling network layer to the target output value.

[0015] Preferably, the arrangement of the physical information network layer is set according to the working arrangement of the turbine hydraulic dynamometer system, and is set in the following order: time information network layer, air intake pipeline network layer, turbine network layer, front bearing network layer, water system network layer, rear bearing network layer, and shaft dynamometer system network layer.

[0016] Preferably, the time information network layer is composed of time term functions, which process current time information and time change information.

[0017] Preferably, the input parameters of the intake pipeline network layer are the parameters collected by the intake pipeline in the actual test system, including intake temperature and intake pressure;

[0018] The input parameters of the turbine network layer are turbine parameters collected by sensors deployed in the turbine components, including inlet and outlet temperatures, inlet and outlet pressures, turbine speed, turbine shaft lubricating oil temperature, and turbine shaft lubricating oil pressure.

[0019] The input parameters of the front bearing network layer are the front bearing parameters of the hydraulic dynamometer, including bearing temperature, bearing oil temperature and oil pressure.

[0020] The input parameters of the water system network layer include inlet and outlet flow rates, inlet and outlet valve openings, and inlet and outlet temperatures.

[0021] The input parameters of the rear bearing network layer are the rear bearing parameters at the rear end of the hydraulic dynamometer, including bearing temperature, bearing oil temperature and oil pressure.

[0022] The input parameters of the shaft power system network layer include shaft power and shaft torque.

[0023] Preferably, the physical information fusion relationship of each sub-network layer in the physical information network layer adopts a bidirectional temporal network structure.

[0024] Preferably, the target output value is torque or power, or front and rear bearing temperatures, or drainage flow rate and drainage temperature.

[0025] Preferably, the physical information network layer further includes a turbine system coupling network layer and a dynamometer system coupling network layer. The output ends of the intake pipe network layer and the turbine network layer are connected to the turbine system coupling network layer. The output ends of the front bearing network layer, the water system network layer, and the rear bearing network layer are connected to the dynamometer system coupling network layer. The output ends of the turbine system coupling network layer and the dynamometer system coupling network layer are connected to the physical information coupling network layer.

[0026] Preferably, the physical information network layer further includes a shaft connection synchronization network layer, which is located between the turbine system coupling network layer and the dynamometer system coupling network layer, and is used to store deviation information between the turbine system and the dynamometer system; the input information of the shaft connection synchronization network layer includes the deviation value between the turbine shaft center and the dynamometer device shaft center.

[0027] Correspondingly: A digital twin model for turbine component testing based on test cumulative effect, including the aforementioned digital twin model for turbine component testing, includes a test cumulative effect network layer and a coupling network layer, wherein the coupling network layer couples the test cumulative effect network layer and the physical information coupling network layer and extracts features; the outputs of the test cumulative effect network layer and the physical information coupling network layer are connected to the coupling network layer.

[0028] Preferably, the test cumulative effect network layer consists of physical information layers at n aforementioned time points. The system performance degradation information is transmitted from time t_0 to time t_1 by a time-series network. The obtained time cumulative information is used as the output of the test cumulative effect network layer, and the output information is input to the coupling network layer.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] 1. By setting up a three-layer structure—Physical Information Network Layer, Physical Information Coupling Network Layer, and Mapping Network Layer—the Physical Information Coupling Network Layer adopts a temporal network structure, connecting the Physical Information Network Layers sequentially in time and space to extract the physical characteristics between the test parameters of the turbine hydraulic dynamometer. This model can perform real-time performance tracking, fault warning, and health management of the test system, improving its safety, stability, and economy. Based on the physical mechanism and system layout of the turbine hydraulic dynamometer, it offers higher accuracy and requires less data compared to purely data-driven modeling methods.

[0031] 2. By setting up three layers—a physical information network layer, a physical information coupling network layer, and a mapping network layer—and simultaneously setting up a test cumulative effect network layer and a coupling network layer, the coupling network layer couples the test cumulative effect network layer and the physical information coupling network layer and extracts features. This model considers the impact of time cumulative effects on the test system during actual operation. By introducing the test cumulative effect layer, performance degradation information and equipment processes are extracted using cumulative effect information learning, which improves the accuracy of the model's performance tracking results by 3-5 times. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of the structure of the turbine component dynamometer device in the background art of this invention;

[0033] Figure 2 This is a schematic diagram of the digital twin model process for turbine component testing according to the present invention;

[0034] Figure 3 This is a schematic diagram of the digital twin model of the turbine system coupling network layer and the dynamometer system coupling network layer introduced in this invention;

[0035] Figure 4 This is a schematic diagram of the digital twin model of the axis connection synchronization network layer introduced in this invention;

[0036] Figure 5 This is a schematic diagram of the bidirectional structure of the physical information network layer of the present invention;

[0037] Figure 6 This is a schematic diagram of the digital twin model for turbine component testing based on the cumulative effect of testing according to the present invention.

[0038] Figure 7 This is a diagram of the internal structure of the cumulative effect network layer in this invention.

[0039] The attached diagram is labeled as follows: rotor 1, coupling 2, housing 3, main shaft 4, base 5. Detailed Implementation

[0040] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Unless otherwise specified, the technical means used in the embodiments are conventional means well known to those skilled in the art.

[0041] like Figure 2 As shown, this invention discloses a digital twin model for turbine component testing based on the working principle of a turbine hydraulic dynamometer. The main architecture of the model includes three layers, S03: a physical information network layer S01, a physical information coupling network layer S02, and a mapping network layer, connected sequentially. The output of the physical information network layer S01 is connected to the physical information coupling network layer S02, and the output of the physical information coupling network layer S02 is connected to the mapping network layer S03.

[0042] The physical information network layer S01, based on the turbine testing system, includes a time information network layer S11, an intake pipe network layer S12, a turbine network layer S13, a front bearing network layer S14, a water system network layer S15, a rear bearing network layer S16, and a shaft power system network layer S17.

[0043] The physical information coupling network layer S02 adopts a temporal network structure, connecting the physical information network layer S01 sequentially in time and space. It extracts the physical correlations and characteristics between the test parameters of the turbine hydraulic dynamometer, such as the correlation features between rotational speed, discharge temperature, discharge valve, etc., and shaft power, and stores these characteristics in the neural network. It should be noted that the input to the physical information coupling network layer S02 is the output of each sub-network of the physical information network layer S01, and the number of input network nodes in the physical information coupling network layer S02 is equal to the number of sub-networks in the physical information network layer S01.

[0044] The mapping network layer S03 maps the information from the physical information coupling network layer S02 to the final target output value. The mapping condition is the mapping relationship of the fully connected layers of the neural network. The number of input nodes of the mapping network layer S03 is equal to the number of output nodes of the physical information coupling network layer S02, and its specific value may vary depending on the actual amount of learning data. Preferably, the number of output nodes of the physical information coupling network layer S02 is greater than its number of input nodes, and the value is a power of 2.

[0045] Furthermore, the arrangement of the physical information network layer S01 is set sequentially according to the working arrangement of the turbine hydraulic dynamometer system, and is set in the following order: time information network layer S11, air intake pipeline network layer S12, turbine network layer S13, front bearing network layer S14, water system network layer S15, rear bearing network layer S16, and shaft dynamometer system network layer S17.

[0046] Furthermore, the time information network layer S11 is composed of time term functions, mainly processing current time information and time change information. For example, time or time difference and other time-related parameters are extracted as physical features in the time information network layer S11. Because time plays a guiding and calibrating role in the actual operation of the test system, the time information network layer S11 is placed at the forefront. The time term function can be any composite parameter containing time information, such as a time interval function, expressed as f = t(n) - t(n-1). Since the turbine system in the test system is a drive system, and it is mainly composed of two parts: the intake pipe and the turbine components, the intake pipe network layer S12 and the turbine network layer S13 follow immediately after the time information network layer S11.

[0047] Furthermore, the input parameters of the intake pipe network layer S12 are parameters collected from the intake pipe in the actual test system, including intake temperature and intake pressure. The input parameters of the turbine network layer S13 are turbine core parameters collected by sensors deployed in the turbine components, including inlet and outlet temperatures, inlet and outlet pressures, turbine speed, turbine shaft lubricating oil temperature, and turbine shaft lubricating oil pressure.

[0048] Specifically, the subsequent front bearing network layer S14, water system network layer S15, rear bearing network layer S16, and shaft dynamometer system network layer S17 correspond to several major component systems of the turbine hydraulic dynamometer. The foremost layer is the front bearing network layer S14, which corresponds to the front bearing section of the turbine hydraulic dynamometer. Its core input parameters are the front bearing parameters at the front end of the hydraulic dynamometer, including bearing temperature, bearing oil temperature, and oil pressure. For example... Figure 1 The schematic diagram of the dynamometer device shows that the water system network layer S15 is located at the rear, and its input parameters include inlet and outlet flow rates, inlet and outlet valve openings, and inlet and outlet temperatures. The rear bearing network layer S16 corresponds to the rear bearing section at the rear end of the hydraulic dynamometer device, and its input parameters are the rear bearing parameters at the rear end of the hydraulic dynamometer device, including bearing temperature, bearing oil temperature, and oil pressure. Since the front bearing, water system, and rear bearing are all driven by the dynamometer shaft, the dynamometer speed is simultaneously used as the input for the front bearing network layer S14, water system network layer S15, and rear bearing network layer S16. Because the final output of the entire testing system is shaft power, the last network layer is set as the shaft power network layer S17, and the input parameters of the shaft power system network layer S17 include shaft power and shaft torque. The arrangement of the physical information network layer S01 is set according to the working arrangement of the turbine-hydraulic dynamometer system, so that the model architecture perfectly matches the physical testing mechanism and structural layout.

[0049] like Figure 5As shown, to connect the sub-network layers of the physical information network layer S01, the physical information coupling network layer S02 uses component parameters along the axis as a digital medium to fuse physical knowledge from the test system. By combining the temporally ordered network results, it arranges the component feature parameters in sequence, forming the internal feature fusion relationship of the physical information input layer. Simultaneously, a bidirectional temporal network structure is considered within the physical information fusion relationship. It should be noted that the temporal structure here can be any existing temporal network, including GRU, LSTM, ARIMA, and BP temporal prediction models, to strengthen the interaction between component feature parameters and eliminate the relationship caused by the order of different measurement parameters representing the same component.

[0050] To illustrate the aforementioned bidirectional timing relationship, in terms of priority, the turbine network layer is directly and positively influenced by the intake pipe network layer S12. However, the turbine network layer S13 also transmits a reverse influence to the intake pipe network layer S12. Although the reverse influence is weaker than the positive influence, this bidirectional effect cannot be eliminated. Explaining this from a practical physical perspective, for example, looking at it from front to back, the intake pressure and temperature change the turbine's work output, thus affecting the rotational speed and consequently the shaft power output of the hydraulic system. Changes in the turbine component's state also, to some extent, alter the airflow in the intake pipe. Simultaneously, the shaft power value acts as a feedback control variable, adjusting the intake parameters to ultimately achieve the rated operating state of the shaft power.

[0051] Furthermore, the target output value can be any required monitoring parameter. Here, core parameters can be taken as monitoring values, such as torque or power as core output performance monitoring values ​​to monitor the overall test performance output; front and rear bearing temperatures as bearing health status monitoring values; and drainage flow rate and drainage temperature as dynamometer heat transfer monitoring parameters.

[0052] Furthermore, such as Figure 3As shown, the physical information network layer S01 also includes a turbine system coupling network layer S101 and a dynamometer system coupling network layer S103. The outputs of the intake pipe network layer S12 and the turbine network layer S13 are connected to the turbine system coupling network layer S101, where the intake pipe network layer S12 and the turbine network layer S13 extract the physical characteristics of the entire turbine system. The outputs of the front bearing network layer S14, the water system network layer S15, and the rear bearing network layer S16 are connected to the dynamometer system coupling network layer S103, where the front bearing network layer S14, the water system network layer S15, and the rear bearing network layer S16 extract the features of the entire turbine-hydraulic dynamometer system within the dynamometer system coupling network layer S103. The outputs of the turbine system coupling network layer S101 and the dynamometer system coupling network layer S103 are connected to the physical information coupling network layer S02. The feature extraction methods of the turbine system coupling network layer S101 and the dynamometer system coupling network layer S103 are similar to those of the physical information coupling network layer S02.

[0053] Furthermore, such as Figure 4 As shown, the physical information network layer S01 also includes a shaft connection synchronization network layer S102. This shaft connection synchronization network layer S102 is located between the turbine system coupling network layer S101 and the dynamometer system coupling network layer S103, and is used to store deviation information between the turbine system and the dynamometer system. The input information for the shaft connection synchronization network layer S102 is the deviation value between the turbine shaft center and the dynamometer equipment shaft center. During the use of the dynamometer system, there will be a deviation between the turbine shaft and the dynamometer equipment shaft center. The main purpose of setting up the shaft connection synchronization network layer S102 is to extract the deviation characteristics and correct the deviation.

[0054] like Figure 6 , 7 As shown, this invention also discloses a digital twin model for turbine component testing based on test cumulative effects. The digital twin model for turbine component testing includes a test cumulative effect network layer S104 and a coupling network layer S202. The coupling network layer S202 couples the test cumulative effect network layer S104 and the physical information coupling network layer S02 and extracts features. The extracted features are the temporal and spatial operational physical correlation features between components and subsystems, i.e., the mutual influence relationships between components in space and time. The outputs of the test cumulative effect network layer S104 and the physical information coupling network layer S02 are connected to the coupling network layer S202.

[0055] Furthermore, the test cumulative effect network layer S104 is composed of n physical information network layers at the aforementioned time points. The system performance degradation information is transmitted from upstream (time t_0) to downstream (time t_1) by the time sequence network, and the obtained time accumulation information is used as the output of the test cumulative effect network layer S104.

[0056] Specifically, the test cumulative effect network layer S104 obtains the system's time accumulation and decay information through the physical information layer at n aforementioned time points. For example, the calculation at time n+k carries the accumulated information from time k-1 to n+k-1, while the calculation at time n+k-1 carries the accumulated information from time k-2 to n+k-2, and so on. Therefore, the actual calculation at time n+k includes the accumulated information from the beginning of the system's operation to the present.

[0057] Specifically, the value of n can be arbitrary, but it should not be too large, otherwise it may cause information overload. Furthermore, since the passage of time is irreversible, the network layer S104 for testing cumulative effects preferably has a unidirectional temporal structure. In addition, the input to the physical information network layer S01 from time t_0 to time t_1 preferably uses parameters strongly correlated with the target value, such as the target value itself.

[0058] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Any modifications, alterations, substitutions, or variations made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention shall fall within the protection scope defined by the claims of the present invention.

Claims

1. A digital twin model for turbine component testing, characterized by: The model architecture body comprises three layers of physical information network layer, physical information coupling network layer and mapping network layer connected in sequence, the output end of the physical information network layer is connected with the physical information coupling network layer, and the output end of the physical information coupling network layer is connected with the three layers of mapping network layer; The physical information network layer comprises time information network layer, air inlet pipeline network layer, turbine network layer, front bearing network layer, water system network layer, rear bearing network layer and shaft power system network layer; The physical information coupling network layer adopts a time sequence network structure, connects the physical information of the physical information network layer through the sequential order of time space, extracts the physical correlation and physical characteristics among the turbine hydraulic dynamometer system test parameters, and stores them to the neural network; The mapping network layer maps the information of the physical information coupling network layer to a target output value; The arrangement of the physical information network layer is sequentially set according to the working arrangement of the turbine hydraulic dynamometer system, and is sequentially set as time information network layer, air inlet pipeline network layer, turbine network layer, front bearing network layer, water system network layer, rear bearing network layer and shaft power system network layer; The time information network layer is composed of time item functions, and processes current time information and time change information; The input parameters of the air inlet pipeline network layer are the parameters collected by the air inlet pipeline in the actual test system, including air inlet temperature and air inlet pressure; The input parameters of the turbine network layer are the turbine parameters collected by the sensors arranged in the turbine components, including inlet and outlet temperature, inlet and outlet pressure, turbine speed, turbine shaft internal oil temperature and turbine shaft internal oil pressure; The input parameters of the front bearing network layer are the front bearing parameters of the front end of the hydraulic dynamometer equipment, including bearing temperature, bearing oil temperature and oil pressure; The input parameters of the water system network layer include water inlet and outlet flow, inlet and outlet valve opening and water inlet and outlet temperature; The input parameters of the rear bearing network layer are the rear bearing parameters of the rear end of the hydraulic dynamometer equipment, including bearing temperature, bearing oil temperature and oil pressure; The input parameters of the shaft power system network layer include shaft power and shaft torque.

2. The digital twin model for turbine component testing of claim 1, wherein: The physical information fusion relationship of each sub-network layer in the physical information network layer adopts a bidirectional time sequence network structure.

3. The digital twin model for turbine component testing of claim 1, wherein: The target output value is torque or power, or front and rear bearing temperature, or water discharge flow and water discharge temperature.

4. The digital twin model for turbine component testing of claim 1, wherein: The turbine system coupling network layer and the dynamometer system coupling network layer are further arranged in the physical information network layer, the output ends of the air inlet pipeline network layer and the turbine network layer are connected with the turbine system coupling network layer, the output ends of the front bearing network layer, the water system network layer and the rear bearing network layer are connected with the dynamometer system coupling network layer, and the output ends of the turbine system coupling network layer and the dynamometer system coupling network layer are connected with the physical information coupling network layer.

5. The digital twin model for turbine component testing of claim 4, wherein: The shaft connection synchronization network layer is further arranged in the physical information network layer, the shaft connection synchronization network layer is arranged between the turbine system coupling network layer and the dynamometer system coupling network layer, and is used for storing the deviation information between the turbine system and the dynamometer system; and the input information of the shaft connection synchronization network layer includes the deviation value of the turbine shaft center and the dynamometer equipment shaft center.

6. Digital twin model for turbine component testing based on cumulative effects of testing, characterized in that: The digital twin model for testing turbine components according to any one of claims 1-5, comprising a test cumulative effect network layer and a coupling network layer, wherein the coupling network layer information-couples the test cumulative effect network layer and a physical information coupling network layer and extracts features; An output end of the test cumulative effect network layer and the physical information coupling network layer is connected to the coupling network layer.

7. The digital twin model for turbine component testing based on cumulative effects of testing of claim 6, wherein: The test cumulative effect network layer is composed of n physical information layers at different time points, a time sequence network is used to transfer system performance degradation information from t_0 time point to t_1 time point, time cumulative information obtained is taken as an output of the test cumulative effect network layer, and the output information is input to the coupling network layer.

Images