Rotor assembly process control method and system based on digital twinning

Abstract

The application discloses a rotor assembly process control method and system based on digital twinning, and relates to the technical field of rotor assembly. The method comprises the following steps: constructing a target simulation model according to geometric parameters and a set of process parameters corresponding to a rotor assembly interface, inputting obtained assembly process parameters into the target simulation model, and obtaining a pressure field snapshot. According to the pressure field snapshot, the target simulation model is subjected to dimension reduction processing to obtain a target reduced-order model. Temperature values obtained from a rotor assembly production line are input into the target reduced-order model to obtain pressure field distribution and temperature field distribution. According to the pressure field distribution and the temperature field distribution, a risk index of the rotor assembly process is calculated. According to the risk index, assembly parameters are adjusted, and rotor assembly operations are performed according to the adjusted assembly parameters. The embodiment of the application can realize precise assembly of the rotor and guarantee process stability and assembly quality of the rotor assembly.

Description

Technical Field

[0001] This application relates to the field of rotor assembly technology, and in particular to a rotor assembly process control method and system based on digital twins. Background Technology

[0002] Rotor assembly, as a core and critical process in the manufacturing of heavy rotating machinery, achieves precise interference fit through controlled heating of the rotor, impeller, and journal or cooling of the shaft, thereby ensuring the rotor's torque transmission capability, assembly coaxiality, and low-vibration operating quality.

[0003] However, the rotor assembly process in related technologies is usually carried out based on preset temperature windows and time thresholds. The lack of real-time detection during the rotor assembly process can easily lead to assembly failures such as rotor jamming, seizing, and cracking, affecting the assembly quality of the rotor and thus reducing the operational stability of heavy rotating machinery. Summary of the Invention

[0004] In view of this, embodiments of this application provide a rotor assembly process control method and system based on digital twins, which can achieve precise rotor assembly, improve rotor assembly quality, and thus improve the operational stability of heavy rotating machinery.

[0005] In a first aspect, this application provides a rotor assembly process control method based on digital twins, comprising: Obtain assembly process parameters and input them into the target simulation model to obtain a snapshot of the pressure field; Based on the pressure field snapshot, the target simulation model is reduced in dimensionality to obtain a reduced-order target model. The obtained temperature values ​​are input into the target reduced-order model to obtain the pressure field distribution and temperature field distribution; Based on the pressure field distribution and temperature field distribution, calculate the risk indicators of the rotor assembly process; Based on the risk indicators, the assembly parameters are adjusted, and the rotor assembly operation is carried out according to the adjusted assembly parameters.

[0006] Secondly, this application provides a rotor assembly process control system, including a model building module, an index calculation module, a parameter adjustment module, and a rotor assembly module; The model building module is used to reduce the dimensionality of the target simulation model based on the pressure field snapshot output by the target simulation model, thereby obtaining a reduced-order target model. The index calculation module is used to calculate the risk index of the rotor assembly process based on the pressure field distribution and temperature field distribution output by the target reduced-order model; The parameter adjustment module is used to adjust assembly parameters based on risk indicators; The rotor assembly module is used to perform rotor assembly operations according to the adjusted assembly parameters.

[0007] By employing the above technical solutions, this application provides a rotor assembly process control method and system based on digital twins. This method reduces the dimensionality of the model by capturing a snapshot of the pressure field output from the target simulation model. Then, it rapidly outputs accurate pressure and temperature field distributions by fusing the actual collected temperature values ​​with the reduced-order model. This allows for the quantification and calculation of risk indicators in the rotor assembly process. Based on these risk indicators, assembly parameters are adjusted accordingly to guide actual assembly operations. In this way, through dynamic optimization and adjustment of assembly parameters, potential risks in the assembly process can be effectively avoided, achieving precise rotor assembly, ensuring the process stability and assembly quality of the rotor, and ultimately improving the operational stability of heavy rotating machinery such as large compressors and turbines.

[0008] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0009] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This paper shows a schematic diagram of the structure of a rotor assembly process control system provided in an embodiment of this application; Figure 2 A schematic flowchart of a rotor assembly process control method based on digital twin provided in an embodiment of this application is shown. Figure 3 This document illustrates a flowchart of a dimensionality reduction process for a target simulation model provided in an embodiment of this application. Figure 4 This illustration shows a schematic diagram of a rotor assembly process for a pressure center of mass, as provided in an embodiment of this application. Detailed Implementation

[0010] To facilitate the explanation of the embodiments of this application, some technical terms and technical means related to the embodiments of this application, as well as the application scenarios of the embodiments of this application, will be introduced first below.

[0011] As the core component of various rotating machinery, the rotor's assembly precision directly determines the equipment's operational stability, transmission efficiency, operating noise, and service life. The selection and control of assembly parameters are key factors affecting rotor assembly precision. However, existing rotor assembly processes typically employ an experience-based open-loop control mode, relying on preset temperature windows and time thresholds for assembly operations. This lacks real-time quantitative detection and closed-loop control of the multi-physical coupling processes of "thermal-mechanical-contact-geometric" interaction during hot assembly. This can easily lead to assembly failures such as rotor jamming, seizing, and cracking, affecting the rotor's assembly quality and consequently reducing the operational stability and reliability of heavy rotating machinery such as large compressors.

[0012] Therefore, to achieve precise rotor assembly, improve rotor assembly quality, and thus enhance the operational stability of heavy rotating machinery, this application provides a rotor assembly process control method based on digital twins, applicable to rotor assembly scenarios of any heavy rotating machinery. In this method, assembly process parameters are acquired and input into a target simulation model to obtain a pressure field snapshot. Then, based on the pressure field snapshot, the target simulation model is dimensionality-reduced to obtain a target reduced-order model. Next, the acquired temperature values ​​are input into the target reduced-order model to obtain the pressure field distribution and temperature field distribution. Then, based on the pressure field distribution and temperature field distribution, risk indicators for the rotor assembly process are calculated. Finally, based on the risk indicators, the assembly parameters are adjusted, and the rotor assembly operation is performed according to the adjusted assembly parameters.

[0013] In this embodiment, model dimensionality reduction is achieved by capturing a snapshot of the pressure field output from the target simulation model. Then, by integrating the reduced-order model with actual collected temperature values, accurate pressure and temperature field distributions are quickly output. This allows for the quantitative calculation of risk indicators during rotor assembly. Based on these risk indicators, assembly parameters are adjusted accordingly to guide actual assembly operations. This not only achieves precise correlation and quantitative analysis between process parameters, physical field distribution, and assembly risks during rotor assembly, significantly improving the timeliness and accuracy of assembly risk identification, but also effectively avoids potential risks during assembly through dynamic parameter optimization. This ensures precise rotor assembly, guarantees the process stability and assembly quality of rotor assembly, and ultimately improves the operational stability of heavy rotating machinery such as large compressors and turbines.

[0014] Furthermore, since the risk indicators for the rotor assembly process are calculated based on the pressure and temperature field distributions output by the target reduced-order model, and the target reduced-order model is obtained by dimensionality reduction based on the target simulation model, this significantly improves the efficiency of physical field simulation analysis and reduces the resource consumption of simulation calculations. This not only adapts to the real-time requirements of actual assembly and facilitates intelligent and precise control of the rotor assembly process, but also provides the necessary conditions for rapid rotor assembly in the future.

[0015] The present application will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the embodiments of the present application can be combined with each other.

[0016] like Figure 1 As shown in the figure, this application provides a rotor assembly process control system 100, including: a model building module 110, an index calculation module 120, a parameter adjustment module 130, and a rotor assembly module 140. The rotor assembly process control system 100 is built based on digital twin technology. This digital twin technology constructs a dynamic digital mirror of a physical entity in virtual space, and through real-time data synchronization, simulation analysis, and closed-loop control, achieves perception, prediction, and optimization of the entire lifecycle of the physical entity.

[0017] The aforementioned model building module 110 is used to construct the target simulation model. This target simulation model is a thermo-mechanical-contact multi-physics coupled simulation model; that is, it is a multi-physics coupled model including heat conduction, structural mechanics, and contact mechanics. Furthermore, this target simulation model is constructed based on the geometric parameters and process parameters corresponding to the rotor assembly interface. This rotor assembly interface is used to connect the impeller and shaft within the rotor. The geometric parameters may include the interface diameter and / or interface length. For example, the interface diameter may be 800 millimeters (mm), and the interface length may be 600 mm. The set of process parameters may include multiple sets of experimental process parameters. Each set of experimental process parameters may include the heating power ratio, holding time, assembly speed, and surface roughness.

[0018] In some cases, the model building module 110 is also used to perform dimensionality reduction on the target simulation model based on the pressure field snapshot output by the target simulation model, thereby obtaining a target reduced-order model. That is, the target reduced-order model is the model obtained after dimensionality reduction of the target simulation model, and it is used to determine the pressure field distribution and temperature field distribution. The pressure field snapshot is an instantaneous record of the pressure distribution state of each grid point within the standard grid at a certain moment. In other words, this pressure field snapshot is used to characterize the pressure field distribution of each grid point within the standard grid at any given time.

[0019] The aforementioned index calculation module 120 is used to calculate the risk indicators of the rotor assembly process based on the pressure field distribution and temperature field distribution output by the target reduced-order model. These risk indicators may include at least one of the following: pressure centroid, peak pressure, edge engagement score, and force pulsation. The pressure centroid is a comprehensive characteristic point describing the pressure distribution within the standard grid; it is the equivalent point of action obtained by weighted averaging of the pressure at all grid points within the standard grid according to their location and magnitude. Peak pressure is the maximum pressure value appearing in the pressure field. The edge engagement score is a core evaluation indicator used to quantitatively assess the fitting quality of the impeller and shaft against the contact surface within the rotor. It obtains a standardized score by detecting and weighting key geometric parameters such as the engagement gap, contact area, alignment, and engagement depth of the engagement edges, thus intuitively reflecting the precision, sealing, and connection reliability of the edge engagement. Force pulsation refers to the periodic or non-periodic fluctuation of fluid forces and / or mechanical contact forces acting on the surface of components over time.

[0020] The parameter adjustment module 130 is used to adjust the assembly parameters according to the aforementioned risk indicators. The assembly parameters may include the impeller thrust in various directions and / or the impeller rotation angle. The impeller thrust in various directions may include the impeller's axial thrust, horizontal thrust, and vertical thrust.

[0021] The rotor assembly module 140 described above is used to perform rotor assembly operations according to the adjusted assembly parameters. The rotor assembly module 140 may include a controller 141 and an industrial computer 142. The controller 141 can be a programmable logic controller (PLC) or an industrial controller, and is used to receive rotor assembly instructions sent by the industrial computer 142, and to mount the impeller onto the rotor shaft according to the instructions. The industrial computer 142 is used to receive assembly parameters sent by the parameter adjustment module 130, and to execute assembly control decisions based on these parameters, thereby generating rotor assembly instructions. In some embodiments of this application, the controller operates at a frequency of 1 kHz, and the industrial computer operates at a frequency of 10-200 Hz.

[0022] It should be noted that before the industrial computer 142 receives the assembly parameters, the assembly parameters need to be converted into data using a six-degrees-of-freedom (6-DOF) platform. After the assembly parameters are converted, they are sent to the industrial computer 142.

[0023] In some cases, the rotor assembly module 140 may also include a temperature sensor. That is, the rotor assembly module 140 is also used to send the temperature values ​​collected by the temperature sensor to the index calculation module, so that the index calculation module can calculate the risk index of the rotor assembly process.

[0024] Optionally, the rotor assembly module 140 can also support a data recording function. This data recording function involves recording and saving the data generated by the rotor assembly process control system 100, the corresponding timestamps and sequence numbers, and the model parameters of the target reduced-order model. Furthermore, a corresponding quality report is generated after each rotor assembly operation. This quality report may include risk indicators and records of anomalies during the rotor assembly process.

[0025] In one implementation, such as Figure 1 As shown, the rotor assembly process control system 100 can trigger the model building module 110 to perform dimensionality reduction processing on the target simulation model based on the pressure field snapshot output by the target simulation model, thereby obtaining a target reduced-order model. Afterwards, the model building module 110 can execute step a, sending the target reduced-order model to the index calculation module 120.

[0026] Simultaneously, the rotor assembly process control system 100 can trigger the temperature sensors included in the rotor assembly module 140 to collect temperature values. Then, the rotor assembly module 140 can execute step b, sending the temperature values ​​to the index calculation module 120. After receiving the target reduced-order model sent by the model building module 110 and the temperature values ​​sent by the rotor assembly module 140, the index calculation module 120 inputs the temperature values ​​into the target reduced-order model to obtain the pressure field distribution and temperature field distribution. Then, based on the pressure field distribution and temperature field distribution, the index calculation module 120 calculates the risk index of the rotor assembly process. Then, the index calculation module 120 can execute step c, sending the risk index of the rotor assembly process to the parameter adjustment module 130.

[0027] Subsequently, upon receiving the risk indicator from the indicator calculation module 120, the parameter adjustment module 130 adjusts the assembly parameters according to the risk indicator. Then, the parameter adjustment module 130 can execute step d, sending the adjusted assembly parameters to the industrial control computer 142 included in the rotor assembly module 140. After receiving the adjusted assembly parameters from the parameter adjustment module 130, the industrial control computer 142 executes assembly control decisions based on the adjusted assembly parameters and generates rotor assembly instructions. Then, the industrial control computer 142 can execute step e, sending the rotor assembly instructions to the controller 141 included in the rotor assembly module 140. After receiving the rotor assembly instructions from the industrial control computer 142, the controller 141 performs rotor assembly operations according to the rotor assembly instructions.

[0028] It should be noted that the communication and interaction between the various modules within the aforementioned rotor assembly process control system 100 is implemented based on the Open Platform Communications Unified Architecture (OPC-UA). This OPC-UA bus architecture is a cross-platform, cross-vendor, service-oriented general communication architecture in the field of industrial automation, used to achieve seamless data interaction, information integration, and remote access between different modules. Furthermore, data transmission between the various modules within the rotor assembly process control system 100 can be achieved based on high-speed Peripheral Component Interconnect Express (PCIe) interface technology. The PCIe interface has a minimum transmission bandwidth of 5 gigabits per second (Gbps).

[0029] Furthermore, this embodiment provides a rotor assembly process control method based on digital twins, such as... Figure 2 As shown, the method includes: S201, obtain the assembly process parameters and input them into the target simulation model to obtain a snapshot of the pressure field.

[0030] The assembly process parameters mentioned above may include at least one of the following: heating power ratio, heat preservation time, assembly speed, surface roughness, thermal conductivity, and coefficient of friction. Thermal conductivity is a physical quantity characterizing the ability of the impeller inside the rotor to transfer heat through the contact surface when it is in contact with the shaft; it reflects the thermal conductivity of the contact surface. The coefficient of friction is a physical quantity characterizing the magnitude of the frictional force between the impeller inside the rotor and the shaft during operation; it reflects the frictional characteristics of the contact surface.

[0031] It is understandable that the material parameters of the impeller and shaft within the rotor are all temperature-dependent. These material parameters can be the elastic modulus, coefficient of thermal expansion, and thermal conductivity. The elastic modulus is an inherent parameter characterizing the impeller and shaft's resistance to elastic deformation. The coefficient of thermal expansion characterizes the impeller and shaft's ability to undergo linear deformation with temperature changes. Thermal conductivity is an inherent parameter characterizing the impeller and shaft's ability to transfer heat.

[0032] Specifically, after obtaining the aforementioned assembly parameters, these assembly process parameters can be input into the target simulation model to obtain a pressure field snapshot. The target simulation model is a thermo-mechanical-contact multiphysics coupled simulation model; that is, it is a multiphysics coupled model including heat conduction, structural mechanics, and contact mechanics. The target simulation model is used to generate simulation datasets covering typical and boundary conditions. The pressure field snapshot is an instantaneous record of the pressure distribution state at each grid point within the specification grid at a given moment. In other words, this pressure field snapshot is used to characterize the pressure field distribution at each grid point within the specification grid at any given time.

[0033] In some cases, inputting the above assembly process parameters into the target simulation model can also yield axial friction force, impeller deformation, and surface temperature across the entire field.

[0034] The aforementioned target simulation model can be constructed based on the geometric parameters and process parameters corresponding to the rotor assembly interface. This rotor assembly interface is used to connect the impeller and shaft within the rotor. The geometric parameters may include the interface diameter and / or interface length. For example, the interface diameter may be 800 millimeters (mm), and the interface length may be 600 millimeters. The set of process parameters may include multiple sets of experimental process parameters. Each set of experimental process parameters may include the heating power ratio, holding time, assembly speed, and surface roughness.

[0035] Specifically, the training process of the aforementioned target simulation model may include: acquiring a set of process parameters; then, for each set of experimental process parameters included in the process parameter set, inputting the experimental process parameters into the pre-constructed simulation model; in response to the completion of training data input, using the least squares method to calibrate the thermal conductivity and friction coefficient of the thermal contact, until the training variables meet the preset conditions, thus obtaining the target simulation model. The training variables may include at least one of the following: temperature values, contact pressure, axial displacement, and axial force for each channel.

[0036] In some cases, where the training variables include temperature values ​​for each channel, the preset condition may include that the root mean square error (RMSE) corresponding to the temperature values ​​for each channel is less than or equal to a preset temperature. The preset temperature can be set in advance according to actual conditions. For example, the preset temperature could be 5 degrees Celsius (°C).

[0037] In other cases, where the training variables include contact pressure, the preset conditions may include a ratio between the contact pressure and the theoretical contact pressure that is less than or equal to a first preset ratio, and / or, the contact pressure being within a preset pressure range. Both the first preset ratio and the preset pressure range can be preset according to actual conditions. For example, the first preset ratio could be 10%.

[0038] In other cases, where the training variables include axial displacement, the preset conditions may include the ratio between the axial displacement and the theoretical axial displacement being less than or equal to a second preset ratio, and / or the axial displacement being within a preset displacement range, and / or the axial displacement being less than a preset displacement. The second preset ratio, the preset displacement range, and the preset displacement can all be preset according to actual conditions.

[0039] In some other cases, where the training variables include axial force, the preset conditions may include a ratio between the axial force and the theoretical axial force that is less than or equal to a third preset ratio, and / or, the axial force being within a preset force range, and / or, the axial force being less than a preset force. The third preset ratio, the preset force range, and the preset force can all be preset according to actual conditions. Furthermore, the third preset ratio, the second preset ratio, and the first preset ratio can be the same or different; no specific limitation is imposed.

[0040] It should be noted that the training process of the above-mentioned target simulation model is based on Orthogonal Experimental Design (OED) experiments. OED experiments are an efficient and economical multi-factor, multi-level experimental design method based on orthogonal arrays, which can significantly reduce the number of experiments and lower costs while ensuring experimental accuracy.

[0041] S202, based on the pressure field snapshot, the target simulation model is dimensionality reduced to obtain the target reduced-order model.

[0042] Specifically, after obtaining the aforementioned pressure field snapshot, the target simulation model can be dimensionality reduced based on this pressure field snapshot.

[0043] In one implementation, such as Figure 3 As shown, the dimensionality reduction process of the above target simulation model can specifically include: S2021, perform singular value decomposition on the pressure field snapshot to obtain the singular value matrix, and extract the singular value vector that conforms to the preset rules from the singular value matrix as a low-dimensional linear term.

[0044] The aforementioned preset rule can be the singular value vector corresponding to the maximum number of singular values ​​in the singular value matrix.

[0045] It is understandable that, since the singular value vectors in the singular value matrix are arranged in descending order of singular value magnitude, a predetermined number of singular value vectors at the beginning of the singular value matrix can be directly selected as low-dimensional linear terms. The predetermined number can be set in advance according to actual needs. For example, the predetermined number can be 50, 60, etc., without any specific limitation.

[0046] S2022, based on the discrete empirical interpolation method, the nonlinear terms included in the target simulation model are reduced in dimensionality to obtain low-dimensional nonlinear terms.

[0047] In this application, the low-dimensional nonlinear terms are adapted to the low-dimensional linear space constructed from the low-dimensional linear terms. In other words, this application uses the low-dimensional linear space constructed from the low-dimensional linear terms as a basis to perform dimensionality reduction processing on the nonlinear terms included in the target simulation model.

[0048] In this embodiment, considering that extracting singular value vectors from the singular value matrix as low-dimensional linear terms only reduces the order of the model's state space, it fails to address the issues of high-dimensional nonlinear terms and high computational cost after order reduction. Therefore, to achieve efficient order reduction across all dimensions of the target simulation model, a discrete empirical interpolation method can be used to reduce the dimensionality of the nonlinear terms in the target simulation model, thereby obtaining low-dimensional nonlinear terms. This not only preserves the simulation accuracy of the target simulation model but also significantly improves computational efficiency.

[0049] Among them, the Discrete Empirical Interpolation Method (DEIM) is used to solve the problem that the dimensionality of the reduced-order model is still too high in the calculation of nonlinear terms. By using empirical interpolation, a small number of representative interpolation points are selected from the high-dimensional discrete data. The low-dimensional information of the interpolation points is used to approximate the high-dimensional nonlinear terms, thereby realizing the secondary dimensionality reduction and accelerated calculation of the reduced-order model.

[0050] S2023. Determine the initial reduced-order model based on the low-dimensional linear and low-dimensional nonlinear terms.

[0051] Specifically, after obtaining the low-dimensional nonlinear term that fits the low-dimensional linear space constructed from the low-dimensional linear term, the initial order reduction model can be determined based on the low-dimensional nonlinear term and the low-dimensional linear term.

[0052] In some cases, according to the model structure of the target simulation model described above, the low-dimensional nonlinear terms are fused with the low-dimensional linear terms to obtain an initial reduced-order model.

[0053] S2024. Based on the residual neural network, the initial order reduction model is corrected to obtain the target order reduction model.

[0054] Specifically, after obtaining the initial reduced-order model, a residual neural network can be used to modify it. This modification can be a physical consistency correction, which ensures that the reduced-order model maintains low-dimensional and efficient solution while conforming to the physical conservation laws, boundary constraints, and constitutive relations of the target simulation model. This reduces the likelihood of simulation results that match mathematical calculations but violate physical reality. For example, violations of physical reality could include negative contact pressure, failure to maintain force balance, or non-conservation of heat flux.

[0055] In some cases, to ensure that the reduced-order model meets the actual rotor assembly requirements, a non-negativity constraint can be added to the loss function corresponding to the reduced-order model. This ensures that the loss value output by the loss function is greater than or equal to 0. This aligns with the physical meaning of the loss, the mathematical logic of optimization, and the actual needs of rotor assembly, reducing the possibility of meaningless negative loss values ​​interfering with model training and disrupting the optimization logic. Simultaneously, the magnitude of the loss value directly reflects the degree of error in the reduced-order model's predictions, providing the necessary conditions for subsequent accurate output of pressure and temperature field distributions.

[0056] S203, input the obtained temperature values ​​into the target reduced-order model to obtain the pressure field distribution and temperature field distribution.

[0057] Specifically, after obtaining the target reduced-order model, temperature values ​​can be collected using temperature sensors. These temperature sensors are installed on the rotor assembly line, meaning the temperature values ​​are obtained from sensors on the rotor assembly line, and these values ​​include the temperature values ​​of each channel. These temperature values ​​can then be input into the target reduced-order model to obtain the pressure field distribution and temperature field distribution. In this way, calculating the pressure field distribution and temperature field distribution using the target reduced-order model can significantly reduce the solution cost of the high-order simulation model while ensuring rotor assembly accuracy, achieving millisecond-level rapid solution of the field distribution, and providing a foundation for subsequent rapid adjustment of assembly parameters.

[0058] In some cases, if no temperature value is collected, or if the collected temperature value is abnormal, linear interpolation or nearest neighbor filling methods can be used to determine the corresponding temperature value, which will then be recorded in the subsequent quality report. Abnormal data can be caused by human error, sensor malfunction, or communication interaction issues.

[0059] S204, calculate the risk index of the rotor assembly process based on the pressure field distribution and temperature field distribution.

[0060] The aforementioned risk indicators may include at least one of the following: pressure centroid, peak pressure, edge engagement score, and force pulsation. The pressure centroid is a comprehensive characteristic point describing the pressure distribution within the specification grid. It is the equivalent point of action obtained by weighting the pressure of all grid points within the specification grid according to their location and magnitude. Peak pressure is the maximum pressure value appearing in the pressure field. The edge engagement score is a core evaluation indicator used to quantitatively assess the fitting quality of the impeller and shaft contact surfaces within the rotor. It obtains a standardized score by detecting and weighting key geometric parameters such as the engagement gap, contact area, alignment, and engagement depth of the engagement edges, thus intuitively reflecting the precision, sealing, and connection reliability of the edge engagement. Force pulsation refers to the periodic or non-periodic fluctuation of fluid forces and / or mechanical contact forces acting on the surface of a component over time.

[0061] In one implementation, the aforementioned pressure centroid can be calculated using the following expression: Expression 1; in, The center of mass of the pressure; For the t-th time point, it is located at a regular grid point (θ). i , z j Pressure field distribution on the grid, θ at regular grid points i The z-axis represents the direction of circular movement around the axial direction, and the z-axis represents the direction of regular grid points. j The axial direction; This represents the position vector of a point on a regular grid in the spatial coordinate system. The area of ​​the regular grid points on the contact area is the area where the impeller and the shaft are in direct contact.

[0062] S205, based on the risk indicators, adjust the assembly parameters and perform rotor assembly operations according to the adjusted assembly parameters.

[0063] Specifically, after obtaining the aforementioned risk indicators, the assembly parameters can be adjusted accordingly. These assembly parameters may include the impeller's thrust in various directions and / or its rotation angle. The impeller's thrust in various directions may include its axial thrust, horizontal thrust, and vertical thrust. Then, the rotor assembly operation can be performed according to the adjusted assembly parameters. The rotor assembly operation involves fitting the impeller onto the shaft.

[0064] In one implementation, such as Figure 4 As shown, when the aforementioned risk indicators include the pressure center of mass, the rotor assembly process targeting the pressure center of mass can specifically include: S2051, determine the axial thrust of the impeller based on the current assembly position of the impeller inside the rotor and the target assembly position of the impeller.

[0065] Specifically, the preset assembly trajectory of the impeller is obtained, and the target assembly position corresponding to the current assembly time point is determined from the preset assembly trajectory. Then, the axial thrust of the impeller is determined based on the current assembly position and the target assembly position of the impeller.

[0066] In one implementation, the axial thrust of the impeller can be calculated using the following expression: Expression 2; in, This refers to the axial thrust of the impeller; This is the second proportionality coefficient; The target assembly location for the impeller; This indicates the current assembly position of the impeller; The second differential coefficient; The displacement increment is obtained after differentiating the target assembly position of the impeller. This is the displacement increment obtained by differentiating the current assembly position of the impeller.

[0067] It should be noted that the above-mentioned second proportional coefficient Second differential coefficient The values ​​are determined based on the current assembly stage of the impeller, which can be either the impeller insertion stage or the impeller locking stage. Specifically, when the current assembly stage is the impeller insertion stage, the second proportional coefficient can be Kins, and the second differential coefficient can be Dins. When the current assembly stage is the impeller locking stage, the second proportional coefficient can be Klock, and the second differential coefficient can be Dlock. Kins is greater than Klock, and Dins is greater than Dlock.

[0068] S2052, determine the horizontal and vertical thrust of the impeller based on the pressure center of mass.

[0069] Specifically, after obtaining the aforementioned pressure centroid, it can be decomposed to obtain its horizontal and vertical components. Then, it can be determined whether the horizontal component is less than a first preset component and whether the vertical component is less than a second preset component. The second and first preset components can be preset according to actual conditions. These components can be the same or different; no specific limitation is made. In some embodiments of this application, both the second and first preset components are 0.5 micrometers (μm).

[0070] In some cases, if the horizontal component of the pressure center of mass is less than the first preset component, it indicates that the impeller is close to the reference origin. The reference origin refers to the core reference datum for the pressure center of mass coordinates. This coordinate system includes both the horizontal and vertical components of the pressure center of mass. Therefore, there is no need to adjust the impeller's horizontal thrust, i.e., there is no need to re-determine the impeller's horizontal thrust. However, if the horizontal component of the pressure center of mass is greater than or equal to the first preset component, it indicates that the impeller is far from the reference origin. Therefore, to bring the impeller closer to the reference origin, the impeller's horizontal thrust can be re-determined based on the horizontal component of the pressure center of mass, and the rotor assembly operation can continue according to the new horizontal thrust.

[0071] In one implementation, the horizontal thrust of the impeller can be calculated using the following expression: Expression 3; in, This is the vertical thrust of the impeller; The first proportionality coefficient; The horizontal component of the pressure center of mass, The first differential coefficient; The horizontal velocity is obtained by differentiating the horizontal component of the pressure centroid.

[0072] In other cases, if the vertical component of the pressure center of mass is less than the second preset component, it indicates that the impeller is close to the reference origin. Therefore, there is no need to adjust the vertical thrust of the impeller, i.e., there is no need to re-determine the vertical thrust of the impeller. However, if the vertical component of the pressure center of mass is greater than or equal to the second preset component, it indicates that the impeller is far from the reference origin. Therefore, in order to bring the impeller closer to the reference origin, the vertical thrust of the impeller can be re-determined based on the vertical component of the pressure center of mass and the weight of the impeller, so that the rotor assembly operation can continue according to the new vertical thrust.

[0073] In one implementation, the vertical thrust of the impeller described above can be calculated using the following expression: Expression 4; in, This is the vertical thrust of the impeller; The first proportionality coefficient; The vertical component of the pressure centroid; The first differential coefficient; The vertical velocity is obtained by differentiating the vertical component of the pressure center of mass. This is the weight of the impeller.

[0074] S2053, according to the axial thrust, horizontal thrust and vertical thrust of the impeller, the impeller is mounted on the rotor shaft.

[0075] Specifically, after obtaining the axial thrust, horizontal thrust, and vertical thrust of the impeller—that is, the thrust of the impeller in each direction—the impeller can be mounted on the rotor shaft according to these thrusts. This allows for timely adjustment of the impeller thrust, reducing the occurrence of over- or under-thrust, effectively mitigating potential risks during assembly, achieving precise rotor assembly, ensuring the process stability and assembly quality of the rotor, and ultimately improving the operational stability of heavy rotating machinery such as large compressors and turbines.

[0076] It should be noted that during rotor assembly according to the impeller's axial thrust, horizontal thrust, and vertical thrust, the actual thrust value needs to be collected using a thrust sensor. Then, the acquisition time of this actual thrust value is aligned with the acquisition time of the aforementioned temperature values. Upon completion of the acquisition time alignment, the actual thrust value is compared with the predicted thrust value to assess the accuracy of the predicted thrust application. The predicted thrust value is determined based on the impeller's axial thrust, horizontal thrust, and vertical thrust.

[0077] Specifically, the actual thrust and temperature values ​​are sent to the target buffer for time alignment using a Precision Time Protocol (PTP). The PTP can be IEEE 1588 PTP, a high-precision clock synchronization protocol for network-based telemetry and control systems. It achieves sub-microsecond or nanosecond-level time synchronization in packet networks such as Ethernet through a master-slave clock architecture and hardware timestamps. Furthermore, a sliding window caching mechanism can be applied to the PTP. This sliding window caching mechanism is a dynamic data caching strategy based on a preset length. It divides the buffer space into a continuous window of a preset length. When new data arrives, the window "slides" in the direction of data inflow, automatically discarding expired data outside the window while retaining the latest and valid data within the window. This achieves a balance between limited reuse of buffer space, timeliness control of data, and access efficiency.

[0078] In another implementation, when the aforementioned risk indicators include edge engagement score and force pulsation, the rotor assembly process targeting the pressure center of mass can specifically include: determining whether the edge engagement score is greater than a preset score, and determining whether the force pulsation is greater than a preset force. If the edge engagement score is greater than the preset score, and / or the force pulsation is greater than the preset force, a bounded sinusoidal disturbance can be activated to suppress stick-slip between the impeller and the shaft. Simultaneously, the rotation angle of the impeller within the rotor is calculated, and the impeller is fitted onto the shaft according to this rotation angle. If the edge engagement score is less than or equal to the preset score, and the force pulsation is less than or equal to the preset force, there is no need to adjust the impeller rotation angle, i.e., there is no need to recalculate the impeller rotation angle.

[0079] Accordingly, the rotation angle of the impeller can be calculated using the following expression: Expression 5; in, Let t be the impeller rotation angle at time t; A is the rotation coefficient, which is less than or equal to 0.05 degrees (°); f is the preset frequency, which can be preset according to actual conditions. For example, the preset frequency f can be 5 Hz.

[0080] Optionally, the assembly control mode can be determined before adjusting the above assembly parameters. The assembly parameters are only adjusted based on risk indicators when the assembly control mode is in the mitigation control mode. The assembly control mode can be a stop control mode, mitigation control mode, warning control mode, or safety control mode.

[0081] In some embodiments, the assembly control mode can be determined by comparing the execution time of the current assembly stage with a preset time. Specifically, if the execution time of the current assembly stage exceeds the preset time, the assembly control mode can be updated to a stop control mode, and the rotor assembly operation can be stopped. If the execution time of the current assembly stage does not exceed the preset time, there is no need to update the assembly control mode.

[0082] In other embodiments, the assembly control mode can be determined by comparing the aforementioned risk indicator with multiple preset thresholds corresponding to the risk indicator. These preset thresholds can be a first preset threshold, a second preset threshold, and a third preset threshold. The first preset threshold is greater than the second preset threshold, and the second preset threshold is greater than the third preset threshold.

[0083] In some cases, taking the aforementioned risk indicator as an example, if the risk indicator is greater than or equal to the first preset threshold, it indicates a significant risk in the rotor assembly process. Therefore, the assembly control mode can be updated to a stop control mode, and the rotor assembly operation can be stopped. This enables timely identification and rapid intervention of potential risks during the assembly process. It not only effectively avoids problems such as substandard rotor assembly accuracy, component damage, and assembly equipment failure caused by continuous assembly with abnormal risk indicators, ensuring the process quality and product yield of rotor assembly, but also reduces rework and repair costs caused by abnormal assembly, improving the overall operational efficiency and production safety of rotor assembly. From the process execution perspective, it provides reliable risk control guarantees for the standardized and regulated implementation of rotor assembly, ensuring that the performance and operational stability of the assembled rotor meet design requirements.

[0084] If the aforementioned risk indicator is less than the first preset threshold, and the risk indicator is greater than or equal to the second preset threshold, it indicates that although there is a significant risk in the rotor assembly process, it is insufficient to cause rotor assembly abnormalities. Therefore, to reduce the probability of rotor assembly abnormalities, the assembly control mode can be updated to a mitigation control mode, and the assembly parameters can be adjusted in a timely manner based on the risk indicator. This can improve rotor assembly accuracy, ensure the process stability and assembly quality of rotor assembly, and thus improve the operational stability of heavy rotating machinery such as large compressors and turbines.

[0085] If the aforementioned risk indicator is less than the second preset threshold, and the risk indicator is greater than or equal to the third preset threshold, it indicates that there is a minor risk in the rotor assembly process. Therefore, the assembly control mode can be updated to a warning control mode, and an anomaly prompt operation can be performed. The anomaly prompt operation can include at least one of the following: playing an alarm sound, displaying an anomaly prompt message on the display screen, displaying a bright red or other colored interface on the display screen, and outputting an anomaly prompt message via voice. Simultaneously, the rotor assembly operation continues according to the original assembly parameters, meaning that the assembly parameters are not adjusted. This allows for timely reminders to staff to focus on key areas, thereby reducing the possibility of rotor assembly anomalies, maximizing rotor assembly accuracy, ensuring the process stability and assembly quality of the rotor assembly, and ultimately improving the operational stability of heavy rotating machinery such as large compressors and turbines.

[0086] If the aforementioned risk indicators are below the third preset threshold, it indicates that there is no risk in the rotor assembly process. Therefore, the assembly control mode can be updated to the safety control mode, and the rotor assembly operation can continue according to the original assembly parameters, meaning that the assembly parameters will not be adjusted. In this way, the stability and continuity of the assembly parameters can be maintained, effectively avoiding problems such as fluctuations in assembly accuracy and decreased process consistency caused by meaningless parameter adjustments, ensuring the process stability and operational efficiency of rotor assembly, and further improving the level of intelligent and precise control of the rotor assembly process.

[0087] In other cases, taking the aforementioned risk indicators as examples, if any of these risk indicators is greater than or equal to a first preset threshold, it indicates a significant risk in the rotor assembly process. Therefore, the assembly control mode can be updated to a stop control mode, and the rotor assembly operation can be stopped. This enables timely identification and rapid intervention of various potential risks during the assembly process. It not only effectively avoids problems such as substandard rotor assembly accuracy, component damage, and assembly equipment failure caused by continuous assembly with abnormal risk indicators, ensuring the process quality and product yield of rotor assembly, but also reduces rework and repair costs caused by abnormal assembly, improving the overall operational efficiency and production safety of rotor assembly. From the process execution perspective, it provides reliable risk control guarantees for the standardized and regulated implementation of rotor assembly, ensuring that the performance and operational stability of the assembled rotor meet design requirements.

[0088] If all the aforementioned risk indicators are below the first preset threshold, but any one of the risk indicators is greater than or equal to the second preset threshold, it indicates that while the rotor assembly process carries significant risk, it is insufficient to cause an abnormality in the rotor assembly. Therefore, to reduce the probability of rotor assembly abnormalities, the assembly control mode can be updated to a mitigation control mode, and the assembly parameters can be adjusted promptly based on the risk indicator. This improves rotor assembly accuracy, ensures the process stability and assembly quality of the rotor assembly, and ultimately enhances the operational stability of heavy rotating machinery such as large compressors and turbines.

[0089] If all the aforementioned risk indicators are below the second preset threshold, but any one of these risk indicators is greater than or equal to the third preset threshold, it indicates a minor risk in the rotor assembly process. Therefore, the assembly control mode can be updated to a warning control mode, and an anomaly alert operation can be initiated. This anomaly alert operation can include at least one of the following: playing an alarm sound, displaying an anomaly alert message on the screen, displaying a bright red or other colored interface on the screen, or outputting an anomaly alert message via voice. Simultaneously, the rotor assembly operation continues according to the original assembly parameters, meaning the assembly parameters are not adjusted. This allows for timely reminders for staff to conduct focused checks, thereby reducing the possibility of rotor assembly anomalies, maximizing rotor assembly accuracy, ensuring the process stability and assembly quality of the rotor assembly, and ultimately improving the operational stability of heavy rotating machinery such as large compressors and turbines.

[0090] If all the aforementioned risk indicators are below the third preset threshold, it indicates that there is no risk in the rotor assembly process. Therefore, the assembly control mode can be updated to the safety control mode, and the rotor assembly operation can continue according to the original assembly parameters, meaning that the assembly parameters will not be adjusted. In this way, the stability and continuity of the assembly parameters can be maintained, effectively avoiding problems such as fluctuations in assembly accuracy and decreased process consistency caused by meaningless parameter adjustments, ensuring the process stability and operational efficiency of rotor assembly, and further improving the level of intelligent and precise control of the rotor assembly process.

[0091] This application also provides a computer device, specifically a personal computer, server, network device, etc. The computer device includes a bus, processor, memory, and communication interface, and may also include input / output interfaces and a display device. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device stores location information. The network interface of the computer device is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements the steps in the various method embodiments.

[0092] Those skilled in the art will understand that the structure of the computer device described above is only a partial structure related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. A specific computer device may include more or fewer components, or combine certain components, or have different component arrangements.

[0093] In one embodiment, a computer-readable storage medium is provided, which may be non-volatile or volatile, having stored thereon a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0094] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0095] It should be noted that the user personal information involved in the embodiments of this application is all authorized (with the knowledge and consent) by the relevant parties or fully authorized by all parties, and the executing entity can obtain it through various legal and compliant means. The collection, storage, use, processing, transmission, provision, and disclosure of the information, data, and signals involved all comply with the relevant laws and regulations of the relevant countries and regions, and do not violate public order and good morals. It should be noted that if any software tools or components other than those of this company appear in the embodiments of this application, they are merely illustrative examples and do not represent actual use.

[0096] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, graphics processors, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0097] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0098] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A rotor assembly process control method based on digital twin, characterized in that, include: The assembly process parameters are obtained and input into the target simulation model to obtain a pressure field snapshot. The assembly process parameters include at least one of heating power ratio, heat preservation time, assembly speed, surface roughness, thermal conductivity and friction coefficient. The target simulation model is a thermo-mechanical-contact multi-physics coupled simulation model, which is constructed based on the geometric parameters and process parameter set corresponding to the rotor assembly interface. Based on the pressure field snapshot, the target simulation model is dimensionality reduced to obtain a reduced-order target model; The obtained temperature values ​​are input into the target reduced-order model to obtain the pressure field distribution and temperature field distribution; Based on the pressure field distribution and the temperature field distribution, the risk indicators of the rotor assembly process are calculated; wherein, the risk indicators include at least one of pressure centroid, peak pressure, edge engagement score, and force pulsation; Based on the aforementioned risk indicators, the assembly parameters are adjusted, and the rotor assembly operation is performed according to the adjusted assembly parameters.

2. The method according to claim 1, characterized in that, The step of adjusting the assembly parameters according to the risk indicators and performing rotor assembly operations according to the adjusted assembly parameters includes: The axial thrust of the impeller is determined based on the current assembly position of the impeller inside the rotor and the target assembly position of the impeller. The horizontal and vertical thrust of the impeller are determined based on the pressure centroid. The impeller is mounted on the rotor shaft according to the axial thrust, horizontal thrust, and vertical thrust of the impeller.

3. The method according to claim 2, characterized in that, The step of determining the horizontal and vertical thrust of the impeller based on the pressure centroid includes: The pressure centroid is decomposed to obtain the horizontal component and the vertical component of the pressure centroid. When the horizontal component of the pressure center of mass is greater than or equal to the first preset component, the horizontal thrust of the impeller is determined based on the horizontal component of the pressure center of mass. When the vertical component of the pressure center of mass is greater than or equal to the second preset component, the vertical thrust of the impeller is determined based on the vertical component of the pressure center of mass and the weight of the impeller.

4. The method according to claim 3, characterized in that, The vertical thrust of the impeller is calculated using the following expression: Expression 1; Among them, the The vertical thrust of the impeller, the The first proportionality coefficient, the The vertical component of the pressure centroid, the The first differential coefficient, the The vertical velocity is obtained by differentiating the vertical component of the pressure centroid. The weight is that of the impeller.

5. The method according to claim 2, characterized in that, The axial thrust of the impeller is calculated using the following expression: Expression 2; Among them, the The axial thrust of the impeller, the The second proportionality coefficient, the The target assembly position of the impeller, the The current assembly position of the impeller, the The second differential coefficient, the The displacement increment obtained by differentiating the target assembly position of the impeller is... The displacement increment is obtained by differentiating the current assembly position of the impeller.

6. The method according to any one of claims 1-5, characterized in that, The step of adjusting the assembly parameters according to the risk indicators and performing rotor assembly operations according to the adjusted assembly parameters further includes: If the edge engagement score is greater than a preset score, and / or the force pulsation is greater than a preset force, calculate the rotation angle of the impeller inside the rotor. According to the rotation angle of the impeller, the impeller is fitted onto the shaft of the rotor.

7. The method according to any one of claims 1-5, characterized in that, When there are multiple risk indicators, adjusting the assembly parameters according to the risk indicators includes: If multiple risk indicators are all less than a first preset threshold, and any one of the multiple risk indicators is greater than or equal to a second preset threshold, the assembly parameters are adjusted according to the multiple risk indicators; wherein the first preset threshold is greater than the second preset threshold.

8. The method according to claim 7, characterized in that, The method further includes: If any of the multiple risk indicators is greater than or equal to the first preset threshold, the rotor assembly operation shall be stopped. If all of the multiple risk indicators are less than the second preset threshold, and any of the multiple risk indicators is greater than or equal to the third preset threshold, an abnormality warning will be issued; wherein, the second preset threshold is greater than the third preset threshold.

9. The method according to any one of claims 1-5, characterized in that, The step of performing dimensionality reduction processing on the target simulation model based on the pressure field snapshot to obtain a target reduced-order model includes: Singular value decomposition is performed on the pressure field snapshot to obtain a singular value matrix, and singular value vectors that conform to preset rules are extracted from the singular value matrix as low-dimensional linear terms; According to the discrete empirical interpolation method, the nonlinear terms included in the target simulation model are subjected to dimensionality reduction processing to obtain low-dimensional nonlinear terms; wherein, the low-dimensional nonlinear terms are adapted to the low-dimensional linear space constructed by the low-dimensional linear terms. The initial reduced-order model is determined based on the low-dimensional linear term and the low-dimensional nonlinear term. The initial order reduction model is modified based on the residual neural network to obtain the target order reduction model.

10. A rotor assembly process control system, characterized in that, The rotor assembly process control system is used to perform the method as described in any one of claims 1 to 9; The rotor assembly process control system includes a model building module, an index calculation module, a parameter adjustment module, and a rotor assembly module. The model building module is used to perform dimensionality reduction processing on the target simulation model based on the pressure field snapshot output by the target simulation model, thereby obtaining a target reduced-order model. The index calculation module is used to calculate the risk index of the rotor assembly process based on the pressure field distribution and temperature field distribution output by the target reduced-order model. The parameter adjustment module is used to adjust the assembly parameters according to the risk indicators; The rotor assembly module is used to perform rotor assembly operations according to the adjusted assembly parameters.

Images