Rotor system response prediction method based on digital twin model

Abstract

The embodiment of the invention discloses a rotor system response prediction method and system based on a digital twin model, and the method comprises the steps: building a dynamic model of a target rotor system through a finite element method, and determining the vibration response data of the target rotor system under different working conditions according to the dynamic model; target parameters influencing vibration response of the target rotor system are obtained, and parameter data of the target parameters under different working conditions are obtained; determining a high-simplification model of the parameter data, the rotating speed under different working conditions and the vibration response data through a GA-BPNN model; obtaining a ratio of the vibration response data to the parameter data, and fitting a fitting relationship between the rotating speed and the ratio through an RBF proxy model; obtaining the actual vibration response of the target rotor system, and reversely deducing the predicted value of the target parameter according to the actual vibration response, the fitting relation and the ratio; and updating the highly simplified model according to the predicted value to obtain an updated digital twinborn model, and completing rotor system response prediction based on the updated digital twinborn model.

Description

Technical Field

[0001] The present invention relates to the technical field of rotor system management, and in particular to a rotor system response prediction method based on a digital twin model. Background Art

[0002] The rotor system has the characteristics of variable loads and harsh working conditions, and is widely used in aircraft engines, gas turbines, etc. In order to ensure the safe and stable operation of the rotor system, it is necessary to conduct comprehensive and systematic monitoring, testing and experiments on its various working conditions and different stages of its life cycle. However, the rotor system used in aircraft engines and gas turbines has a compact structure and small space. It is difficult to arrange sensors at key positions on the rotor, and the test is difficult to carry out and measure. This has caused great difficulties in studying its vibration characteristics. The digital twin model is the core means of digital twin technology. It is an integrated model of physical entities in virtual space and a true digital archive of the entire life cycle of the rotor system. By establishing a real-time interactive digital twin of the rotor system and integrating information interaction, data analysis, feature extraction and other methods, a virtual twin model of the physical system can be constructed through limited observation points, which can be used to solve such difficult-to-monitor problems.

[0003] However, in the existing technology, the twin model is relatively complex. It is difficult to accurately predict the response of the difficult-to-measure points in the rotor system through the actual measurement data of the difficult-to-measure points, which seriously affects the accuracy of the digital twin. Summary of the invention

[0004] In view of this, the present invention provides a rotor system response prediction method based on a digital twin model, which is used to solve the problem in the prior art that it is difficult to accurately predict its response through actual measurement data of difficult-to-measure points.

[0005] In a first aspect, an embodiment of the present invention provides a rotor system response prediction method based on a digital twin model, the method comprising:

[0006] Establishing a dynamic model of a target rotor system by a finite element method, and determining vibration response data of the target rotor system under different working conditions according to the dynamic model;

[0007] Determining a target parameter affecting the vibration response of the target rotor system according to the vibration response data, and acquiring parameter data of the target parameter under different working conditions;

[0008] Determine the parameter data, the rotation speed under different working conditions and the highly simplified model of the vibration response data by using the GA-BPNN model;

[0009] Acquire a ratio of the vibration response data to the parameter data, and fit a fitting relationship between the rotation speed and the ratio through an RBF proxy model;

[0010] Acquire an actual rotational speed of the target rotor system and an actual vibration response at the actual rotational speed, and inversely infer a predicted value of the target parameter according to the actual rotational speed, the actual vibration response, the fitting relationship and the ratio;

[0011] The highly refined simplified model is updated according to the predicted value to obtain an updated digital twin model, and the response prediction of the difficult-to-measure points in the rotor system is completed based on the updated digital twin model.

[0012] Optionally, the step of establishing a dynamic model of the target rotor system by using a finite element method, and determining vibration response data of the target rotor system under different working conditions according to the dynamic model includes:

[0013] Acquiring attribute data of the target rotor system;

[0014] According to the attribute data, a motion equation of the target rotor system is obtained by Lagrangian energy method, and the motion equation is used as a dynamic model of the target rotor system;

[0015] The Newmark method is used to solve the motion equation to obtain vibration response data of the target rotor system under different working conditions.

[0016] Optionally, the step of determining a target parameter affecting the vibration response of the target rotor system according to the vibration response data comprises:

[0017] Acquiring a preset parameter set that affects the vibration response of the target rotor system;

[0018] Calculating the initial sensitivity coefficient of each preset parameter in the preset parameter set based on the finite difference method, and performing dimensionless processing on the initial sensitivity coefficient to obtain the target sensitivity coefficient;

[0019] Calculating the system sensitivity of the target rotor system according to the vibration response data and a preset formula;

[0020] A target parameter affecting the vibration response of the target rotor system is determined according to the target sensitivity coefficient and the system sensitivity.

[0021] Optionally, before the step of determining the parameter data, the rotation speed under different working conditions and the highly simplified model of the vibration response data by using the GA-BPNN model, the method further includes:

[0022] The initial weights and initial thresholds in the error back propagation algorithm are updated by the genetic algorithm to obtain an initial error back propagation algorithm;

[0023] The initial error back propagation algorithm is trained until the initial error back propagation algorithm meets a preset condition, thereby obtaining the GA-BPNN model.

[0024] Optionally, the mathematical expression of the RBF proxy model is:

[0025]

[0026] Among them, x is the predicted sample point, x i is the i-th known sample point, y(x) is the predicted value of the predicted sample point, ω is the weight coefficient, dist is the Euclidean distance, RBF is the radial basis function, b is the output layer bias parameter, and specifically, the radial basis function is:

[0027]

[0028] Optionally, the step of inversely deducing the predicted value of the target parameter according to the actual rotation speed, the actual vibration response, the fitting relationship and the ratio includes:

[0029] According to the actual speed, solving the fitting relationship represented by the RBF proxy model by the OLS method to obtain a target ratio corresponding to the actual speed;

[0030] A predicted value of the target parameter is determined based on the target ratio and the actual vibration response.

[0031] On the other hand, a rotor system response prediction system based on a digital twin model comprises:

[0032] A physical subunit, the physical subunit is used to obtain an actual rotational speed of the target rotor system and an actual vibration response at the actual rotational speed;

[0033] An offline subunit, wherein the offline subunit is used to establish a dynamic model of a target rotor system by using a finite element method;

[0034] A digital twin subunit, the digital twin subunit is used to determine the vibration response data of the target rotor system under different working conditions according to the dynamic model; determine the target parameters that affect the vibration response of the target rotor system according to the vibration response data, and obtain the parameter data of the target parameters under different working conditions; determine the parameter data, the rotational speed under different working conditions and the highly simplified model of the vibration response data through the GA-BPNN model; obtain the ratio of the vibration response data and the parameter data, and fit the fitting relationship between the rotational speed and the ratio through the RBF proxy model; obtain the actual rotational speed of the target rotor system and the actual vibration response at the actual rotational speed, and reversely infer the predicted value of the target parameter according to the actual rotational speed, the actual vibration response, the fitting relationship and the ratio; update the highly simplified model according to the predicted value to obtain an updated digital twin model, and complete the response prediction of the difficult-to-measure points in the rotor system based on the updated digital twin model.

[0035] Optionally, the physical subunit and the digital twin subunit use the WebSocket full-duplex communication protocol to exchange data.

[0036] Optionally, the digital twin sub-unit is a visualization unit built based on JavaScript.

[0037] Optionally, the digital twin sub-unit is a data processing unit constructed by jointly programming MATLAB and LabVIEW based on MATLABScript.

[0038] The technical solution of the embodiment of the present invention is to establish a dynamic model of the target rotor system through the finite element method, and determine the vibration response data of the target rotor system under different working conditions according to the dynamic model; determine the target parameters affecting the vibration response of the target rotor system according to the vibration response data, and obtain the parameter data of the target parameters under different working conditions; determine the parameter data, the rotational speed under different working conditions and the highly simplified model of the vibration response data through the GA-BPNN model; obtain the ratio of the vibration response data and the parameter data, and fit the fitting relationship between the rotational speed and the ratio through the RBF proxy model; obtain the actual rotational speed of the target rotor system and the actual vibration response at the actual rotational speed, and reversely infer the predicted value of the target parameter according to the actual rotational speed, the actual vibration response, the fitting relationship and the ratio; update the highly simplified model according to the predicted value to obtain an updated digital twin model, and complete the response prediction of the difficult-to-measure points in the rotor system based on the updated digital twin model. The different correction coefficients of the digital twin model under different working conditions are taken into account to realize the prediction of the responses of difficult-to-measure points under variable working conditions of the rotor system. The vibration signals (actual vibration responses) of the digital twin and the easy-to-measure points of the physical entity are corrected at the corresponding positions. The prediction of the vibration signals of the difficult-to-measure points can restore the vibration signals of the difficult-to-measure positions of the physical entity to the greatest extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] in:

[0041] Figure 1 is a schematic flow chart of a rotor system response prediction method based on a digital twin model in an embodiment;

[0042] Figure 2 is an application scenario diagram of a rotor system response prediction method based on a digital twin model in an embodiment;

[0043] Figure 3 is a schematic diagram of a finite element model of a rotor system response prediction method based on a digital twin model in an embodiment;

[0044] Figure 4 Schematic diagram of a digital twin platform implementation process in a rotor system response prediction method based on a digital twin model in an embodiment;

[0045] Figure 5 is a structural schematic diagram of a rotor system response prediction system based on a digital twin model in one embodiment;

[0046] Figure 6 A WebSocket communication flow chart of a rotor system response prediction system based on a digital twin model in one embodiment;

[0047] Figure 7 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application;

[0048] Figure 8 It is a structural diagram of a computer-readable storage provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] In one embodiment, the present invention provides a rotor system response prediction method based on a digital twin model, and the rotor system response prediction method based on a digital twin model of the present invention can be performed by a rotor system response prediction device based on a digital twin model. The rotor system response prediction device based on a digital twin model can be implemented by software and / or hardware.

[0051] like Figure 1 As shown, the rotor system response prediction method based on the digital twin model of the embodiment of the present invention specifically includes the following steps:

[0052] S110, establishing a dynamic model of a target rotor system by using a finite element method, and determining vibration response data of the target rotor system under different working conditions according to the dynamic model;

[0053] In a possible implementation manner, the step of establishing a dynamic model of the target rotor system by using a finite element method, and determining vibration response data of the target rotor system under different working conditions according to the dynamic model includes:

[0054] Acquiring attribute data of the target rotor system;

[0055] According to the attribute data, a motion equation of the target rotor system is obtained by Lagrangian energy method, and the motion equation is used as a dynamic model of the target rotor system;

[0056] The Newmark method is used to solve the motion equation to obtain vibration response data of the target rotor system under different working conditions.

[0057] For example, the finite element method is used to establish the rotor system dynamics model, such as Figure 3 As shown, the finite element model of the target rotor system has a total of 16 nodes, including an elastic shaft, a rigid disk and a support unit. The shaft model is established using a Timoshenko beam unit with 8 degrees of freedom, the disk is considered as a rigid disk with concentrated mass, and the support unit is simulated using a spring-damper unit that only considers stiffness and damping. The motion equation of the target rotor system is obtained according to the Lagrangian energy method:

[0058]

[0059] In the formula, is the overall displacement vector of the rotor system, is the overall velocity vector of the rotor system, is the overall acceleration vector of the rotor system, M and C are the overall mass matrix and overall damping matrix of the rotor system, G and K are the overall gyroscopic force matrix and overall stiffness matrix of the rotor system, Ω is the rotation speed of the rotor system, It is the external load on the rotor system. and They are all 4n×1-order vectors; n is the number of nodes after the rotor is divided by finite element.

[0060] Exemplarily, the Newmark method is used to solve the second-order differential motion equation (the motion equation of the target rotor system), so as to obtain the vibration response of the rotor system under different working conditions, thereby providing a data source for the establishment of a highly refined and simplified model.

[0061] S120, determining a target parameter that affects the vibration response of the target rotor system according to the vibration response data, and acquiring parameter data of the target parameter under different working conditions;

[0062] For example, the parameters that affect the vibration response of the rotor system mainly include imbalance, damping and support stiffness. Since support stiffness will affect the critical speed, only the sensitivity of damping and imbalance to amplitude is analyzed. In the full speed range, imbalance has an impact on the vibration response of the rotor, while damping has a certain influence only at the critical speed, and the influence is relatively small compared with the imbalance.

[0063] S130, determining a highly simplified model of the parameter data, the rotation speed under different working conditions, and the vibration response data through a GA-BPNN model;

[0064] S140, obtaining a ratio of the vibration response data to the parameter data, and fitting a fitting relationship between the rotation speed and the ratio through an RBF proxy model;

[0065] Exemplarily, the RBF proxy model is a real-valued function whose value depends only on the distance from the origin, using the Euclidean distance as a metric. The model has a simple structure, efficient training and fast convergence speed, can approximate various nonlinear functions, and has a certain degree of interpretability, so the RBF proxy model is used to fit the relationship between the rotational speed and the LR value (the ratio of the vibration response data to the parameter data).

[0066] S150, acquiring an actual rotational speed of the target rotor system and an actual vibration response at the actual rotational speed, and inferring a predicted value of the target parameter according to the actual rotational speed, the actual vibration response, the fitting relationship and the ratio;

[0067] Exemplarily, the digital twin model is simplified. When the speed is constant, the output amplitude of the twin model (the vibration response data) and the input imbalance (the parameter data) can be approximately linearly related. The ratio is set as LR through linear regression analysis. Therefore, it is only necessary to determine the relationship between the speed and the model correction coefficient (the ratio of the twin model output amplitude (the vibration response data) to the input imbalance (the parameter data), LR value) under variable working conditions to complete the dynamic update of the digital twin model under variable working conditions.

[0068] S160. Update the highly refined simplified model according to the predicted value to obtain an updated digital twin model, and complete the response prediction of the difficult-to-measure points in the rotor system based on the updated digital twin model.

[0069] Exemplarily, the vibration signals (actual vibration responses) of the easy-to-measure points of the digital twin and the physical entity are corrected at corresponding positions, and the vibration signals of the difficult-to-measure points are predicted, so that the vibration signals of the difficult-to-measure positions of the physical entity can be restored to the greatest extent.

[0070] In a possible implementation, Figure 2 As shown in the figure, an application scenario diagram of a rotor system response prediction method based on a digital twin model is shown. The rotor system response prediction method based on a digital twin model can be divided into two stages: offline and online. In the offline stage, the finite element method is used to establish the rotor system dynamics model, and sufficient data is generated to train a highly refined simplified model as a GA-BPNN data-driven mirror of the dynamics model (twin data of the physical entity and the twin model interaction); in the online stage, a model dynamic update method based on an RBF proxy model is used to achieve state synchronization between the digital twin and the physical entity and real-time prediction of the response of difficult-to-measure points through real-time information interaction with the physical entity.

[0071] The twin model is the core means of digital twin technology. It is an integrated model of physical entities in virtual space and a real digital archive of the entire life cycle of the aircraft engine rotor system. In order to carry out the response prediction of the difficult-to-measure points of the rotor system, a digital twin model of the rotor system is established, including geometric models, dynamic models and rule models. The geometric model is built (modeled) based on the basic information of the rotor system obtained through the physical entity (part size, shape characteristics, state tracking and assembly relationship). It is a basic model that reflects physical vibration and intuitively displays the real-time operating status to the user; the dynamic model is constructed based on the motion equation. The dynamic model is a mechanism model that reflects the vibration characteristics of the system and is the core of the operation of the digital twin model; the rule model includes a high-precision simplified model and an update model. It extracts rules from historical data and professional knowledge and is the key to realizing the response prediction (amplitude B1, B2) and model update of the digital twin model based on the speed (W1, W2) and imbalance.

[0072] The physical entity includes a rotor system, sensor communication and a status monitoring system. The twin model evaluates and predicts the physical entity, and the physical entity updates the twin model.

[0073] A dynamic model of the target rotor system is established by the finite element method, and the vibration response data of the target rotor system under different working conditions is determined according to the dynamic model; the target parameters affecting the vibration response of the target rotor system are determined according to the vibration response data, and the parameter data of the target parameters under different working conditions are obtained; a highly simplified model of the parameter data, the rotational speed under different working conditions and the vibration response data is determined by the GA-BPNN model; the ratio of the vibration response data and the parameter data is obtained, and the fitting relationship between the rotational speed and the ratio is fitted by the RBF proxy model; the actual rotational speed of the target rotor system and the actual vibration response at the actual rotational speed are obtained, and the predicted value of the target parameter is inferred according to the actual rotational speed, the actual vibration response, the fitting relationship and the ratio; the highly simplified model is updated according to the predicted value to obtain an updated digital twin model, and the response prediction of the difficult-to-measure points in the rotor system is completed based on the updated digital twin model. The different correction coefficients of the digital twin model under different working conditions are taken into account to realize the prediction of the responses of difficult-to-measure points under variable working conditions of the rotor system. The vibration signals (actual vibration responses) of the digital twin and the easy-to-measure points of the physical entity are corrected at the corresponding positions. The prediction of the vibration signals of the difficult-to-measure points can restore the vibration signals of the difficult-to-measure positions of the physical entity to the greatest extent.

[0074] In a possible implementation manner, the step of determining the target parameter affecting the vibration response of the target rotor system according to the vibration response data includes:

[0075] Acquiring a preset parameter set that affects the vibration response of the target rotor system;

[0076] Calculating the initial sensitivity coefficient of each preset parameter in the preset parameter set based on the finite difference method, and performing dimensionless processing on the initial sensitivity coefficient to obtain the target sensitivity coefficient;

[0077] Calculating the system sensitivity of the target rotor system according to the vibration response data and a preset formula;

[0078] A target parameter affecting the vibration response of the target rotor system is determined according to the target sensitivity coefficient and the system sensitivity.

[0079] Exemplarily, the sensitivity coefficient of each parameter is calculated based on the finite difference method and dimensionless processing is performed. The formula for calculating the local sensitivity of the parameter using the finite difference method is:

[0080]

[0081] Where x is the design variable, x0 is the initial value of the design variable, Δx is a small perturbation of the design variable, and y is the output variable.

[0082] The calculation formula of rotor system sensitivity is defined as:

[0083]

[0084] In the formula, S is the sensitivity, P is the vibration response of the rotor system, U0 is the initial value of the influencing parameter, and ΔU is the small perturbation of the influencing parameter. By calculating the sensitivity coefficient, the sensitivity degree affecting the vibration response of the rotor is determined as unbalance > damping > support stiffness, that is, the unbalance is the target parameter.

[0085] In a possible implementation manner, before the step of determining the parameter data, the rotation speed under different working conditions and the highly simplified model of the vibration response data by using the GA-BPNN model, the step further includes:

[0086] The initial weights and initial thresholds in the error back propagation algorithm are updated by the genetic algorithm to obtain an initial error back propagation algorithm;

[0087] The initial error back propagation algorithm is trained until the initial error back propagation algorithm meets a preset condition, thereby obtaining the GA-BPNN model.

[0088] For example, the mapping relationship between speed, imbalance and vibration response is established based on GA-BPNN. As a highly simplified model of the rotor dynamics model, the vibration response can be quickly output under given model parameters. Therefore, the calculation time can be significantly reduced during model update and response prediction. The initial weights and thresholds of the neural network are optimized through genetic algorithms, which improves the accuracy of the BP neural network and ensures the accuracy of the digital twin model.

[0089] In a possible implementation, the mathematical expression of the RBF proxy model is:

[0090]

[0091] Among them, x is the predicted sample point, x i is the i-th known sample point, y(x) is the predicted value of the predicted sample point, ω is the weight coefficient, dist is the Euclidean distance, RBF is the radial basis function, b is the output layer bias parameter, and specifically, the radial basis function is:

[0092]

[0093] In a possible implementation manner, the step of inferring the predicted value of the target parameter according to the actual rotation speed, the actual vibration response, the fitting relationship and the ratio includes:

[0094] According to the actual speed, solving the fitting relationship represented by the RBF proxy model by the OLS method to obtain a target ratio corresponding to the actual speed;

[0095] A predicted value of the target parameter is determined based on the target ratio and the actual vibration response.

[0096] For example, RBF fits the relationship between speed and LR value better than other proxy models, and can predict LR values ​​at different speeds, determine the current imbalance parameter (predicted value of the target parameter) based on the LR value, and thus realize dynamic update of the model under variable operating conditions.

[0097] In a possible implementation, Figure 4As shown in the figure, the implementation process of the digital twin platform in a rotor system response prediction method based on a digital twin model is based on a B / S architecture, and the digital twin platform is developed using technologies such as HTML, CSS, and JavaScript. The LabVIEW operating platform communicates with the device to achieve real-time data acquisition and dynamic model update. The model visualization is developed based on the encapsulation function library three.js. Three.js is a JavaScript framework based on WebGL, which abstracts and encapsulates the WebGL API. The overall steps of three.js are: first set the scene and add light sources, import the model and configure the material, and finally set the camera parameters and use the renderer to complete the scene construction.

[0098] Responsive visualization is developed based on the encapsulated function library echarts.js. echarts.js is a data visualization chart library based on JavaScript that can render actual data and model data. The overall steps of echarts.js are: first prepare a container (dom) of a given size for rendering charts, then initialize the echarts instance object, and finally set the configuration items to the echarts instance object to complete the rendering of the chart.

[0099] The core of the digital twin model is to continuously update the running status of the digital twin model through data interaction between the virtual twin model and the actual physical model. LabVIEW is a graphical programming software that supports a variety of data acquisition hardware and provides tools for data acquisition, processing and analysis. It is an important bridge for realizing the virtual-real interaction of digital twins. A data acquisition system is built based on LabVIEW-NIcDAQ, and real-time data acquisition is realized in combination with NI devices. Data interaction is the key to achieving state synchronization between virtual models and physical entities. The WebSocket communication protocol is a full-duplex communication based on a TCP connection. After a handshake is established, the client and the server can actively request each other, and support persistent connections, providing support for the real-time interaction between virtual models and physical models. LabVIEW is used to establish the server and transmit the collected signals to the client after processing. JavaScript is used to establish the client, receive real-time information from the server, perform visual rendering through echarts, and drive the model through object3D.

[0100] On the other hand, Figure 5 As shown, a rotor system response prediction system based on a digital twin model, the system comprising:

[0101] A physical subunit 201, the physical subunit is used to obtain the actual rotational speed of the target rotor system and the actual vibration response at the actual rotational speed;

[0102] An offline subunit 202, the offline subunit is used to establish a dynamic model of the target rotor system by using a finite element method;

[0103] A digital twin subunit 203, the digital twin subunit is used to determine the vibration response data of the target rotor system under different working conditions according to the dynamic model; determine the target parameters affecting the vibration response of the target rotor system according to the vibration response data, and obtain the parameter data of the target parameters under different working conditions; determine the parameter data, the rotational speed under different working conditions and the highly simplified model of the vibration response data through the GA-BPNN model; obtain the ratio of the vibration response data to the parameter data, and fit the fitting relationship between the rotational speed and the ratio through the RBF proxy model; obtain the actual rotational speed of the target rotor system and the actual vibration response at the actual rotational speed, and reversely infer the predicted value of the target parameter according to the actual rotational speed, the actual vibration response, the fitting relationship and the ratio; update the highly simplified model according to the predicted value to obtain an updated digital twin model, and complete the response prediction of the difficult-to-measure points in the rotor system based on the updated digital twin model.

[0104] In a possible implementation, Figure 6 As shown, the physical subunit and the digital twin subunit use the WebSocket full-duplex communication protocol to exchange data.

[0105] Exemplarily, the interaction process is: create a WebSocket object to pass in the URL and send a link request; create a WebSocket server object to listen to client requests; establish a connection and send information onOpen() method; send a message to the server; the server executes the text() method to access data and send data; the client receives data and triggers onMessage(); update Echarts, and the data is dynamically displayed; check whether the connection is closed, if closed, then end, if not closed, then continue with the step of the client receiving data and triggering onMessage().

[0106] In a possible implementation, the digital twin sub-unit is a visualization unit built based on JavaScript.

[0107] For example, JavaScript is a typical scripting language for developing web pages, which contains a large number of encapsulated function libraries for the visualization of digital twin models.

[0108] In a possible implementation, the digital twin subunit is a data processing unit constructed by jointly programming MATLAB and LabVIEW based on MATLABScript.

[0109] Exemplarily, since the data calculation capability of LabVIEW is limited, MATLAB-LabVIEW joint programming is implemented based on MATLABScript, which improves the model calculation speed and data processing capability.

[0110] In another embodiment of the present invention, an electronic device is also provided. Figure 7 A block diagram of an exemplary electronic device 50 suitable for implementing exemplary embodiments of the present invention is shown.

[0111] like Figure 7 As shown, the electronic device 50 is in the form of a general computing device. The components of the electronic device 50 may include but are not limited to: one or more processors or processing units 501, a system memory 502, and a bus 503 connecting different system components (including the system memory 502 and the processing unit 501).

[0112] Bus 503 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus structures. For example, these architectures include but are not limited to Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus and Peripheral Component Interconnect (PCI) bus.

[0113] The electronic device 50 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the electronic device 50, including volatile and non-volatile media, removable and non-removable media.

[0114] System memory 502 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 504 and / or cache memory 505. Electronic device 50 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 506 may be used to read and write non-removable, non-volatile magnetic media ( Figure 7 not shown, usually called a "hard drive"). Although Figure 7Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 503 via one or more data medium interfaces. The memory 502 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the embodiments of the present invention.

[0115] A program / utility 508 having a set (at least one) of program modules 507 may be stored, for example, in the memory 502, such program modules 507 including but not limited to an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. The program modules 507 generally perform the functions and / or methods of the embodiments described herein.

[0116] The electronic device 50 may also communicate with one or more external devices 509 (e.g., keyboards, pointing devices, displays 510, etc.), may communicate with one or more devices that enable a user to interact with the electronic device 50, and / or communicate with any device that enables the electronic device 50 to communicate with one or more other computing devices (e.g., a network card, a modem, etc.). Such communication may be performed through an input / output (I / O) interface 511. Furthermore, the electronic device 50 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 512. As shown, the network adapter 512 communicates with other modules of the electronic device 50 through the bus 503. It should be understood that although Figure 3 Not shown, other hardware and / or software modules may be used in conjunction with the electronic device 50, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0117] The processing unit 501 executes various functional applications and data processing by running the programs stored in the system memory 502, such as implementing the rotor system response prediction method based on the digital twin model provided in an embodiment of the present invention.

[0118] In another embodiment of the present invention, Figure 8 As shown, a storage medium 400 including a computer program 411 is also provided. When the computer program 411 is executed by a computer processor, it is used to perform a rotor system response prediction method based on a digital twin model, and the method includes:

[0119] Establishing a dynamic model of a target rotor system by a finite element method, and determining vibration response data of the target rotor system under different working conditions according to the dynamic model;

[0120] Determining a target parameter affecting the vibration response of the target rotor system according to the vibration response data, and acquiring parameter data of the target parameter under different working conditions;

[0121] Determine the parameter data, the rotation speed under different working conditions and the highly simplified model of the vibration response data by using the GA-BPNN model;

[0122] Acquire a ratio of the vibration response data to the parameter data, and fit a fitting relationship between the rotation speed and the ratio through an RBF proxy model;

[0123] Acquire an actual rotational speed of the target rotor system and an actual vibration response at the actual rotational speed, and inversely infer a predicted value of the target parameter according to the actual rotational speed, the actual vibration response, the fitting relationship and the ratio;

[0124] The highly refined simplified model is updated according to the predicted value to obtain an updated digital twin model, and the response prediction of the difficult-to-measure points in the rotor system is completed based on the updated digital twin model.

[0125] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, - but not limited to - an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0126] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0127] Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0128] Computer program code for performing the operation of embodiments of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0129] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A rotor system response prediction method based on a digital twin model, characterized in that: include: Establishing a dynamic model of a target rotor system by a finite element method, and determining vibration response data of the target rotor system under different working conditions according to the dynamic model; Determining a target parameter affecting the vibration response of the target rotor system according to the vibration response data, and acquiring parameter data of the target parameter under different working conditions; Determine the parameter data, the rotation speed under different working conditions and the highly simplified model of the vibration response data by using the GA-BPNN model; Acquire a ratio of the vibration response data to the parameter data, and fit a fitting relationship between the rotation speed and the ratio through an RBF proxy model; Acquire an actual rotational speed of the target rotor system and an actual vibration response at the actual rotational speed, and inversely infer a predicted value of the target parameter according to the actual rotational speed, the actual vibration response, the fitting relationship and the ratio; The highly refined simplified model is updated according to the predicted value to obtain an updated digital twin model, and the response prediction of the difficult-to-measure points in the rotor system is completed based on the updated digital twin model.

2. The method according to claim 1, characterized in that The step of establishing a dynamic model of the target rotor system by the finite element method, and determining the vibration response data of the target rotor system under different working conditions according to the dynamic model comprises: Acquiring attribute data of the target rotor system; According to the attribute data, a motion equation of the target rotor system is obtained by Lagrangian energy method, and the motion equation is used as a dynamic model of the target rotor system; The Newmark method is used to solve the motion equation to obtain vibration response data of the target rotor system under different working conditions.

3. The method according to claim 1, characterized in that The step of determining the target parameter affecting the vibration response of the target rotor system according to the vibration response data comprises: Acquiring a preset parameter set that affects the vibration response of the target rotor system; Calculating the initial sensitivity coefficient of each preset parameter in the preset parameter set based on the finite difference method, and performing dimensionless processing on the initial sensitivity coefficient to obtain the target sensitivity coefficient; Calculating the system sensitivity of the target rotor system according to the vibration response data and a preset formula; A target parameter affecting the vibration response of the target rotor system is determined according to the target sensitivity coefficient and the system sensitivity.

4. The method according to claim 1, characterized in that: Before the step of determining the parameter data, the rotation speed under different working conditions and the highly simplified model of the vibration response data by the GA-BPNN model, the method further includes: The initial weights and initial thresholds in the error back propagation algorithm are updated by the genetic algorithm to obtain an initial error back propagation algorithm; The initial error back propagation algorithm is trained until the initial error back propagation algorithm meets a preset condition, thereby obtaining the GA-BPNN model.

5. The method according to claim 1, characterized in that The mathematical expression of the RBF proxy model is: Among them, x is the predicted sample point, x i is the i-th known sample point, y(x) is the predicted value of the predicted sample point, ω is the weight coefficient, dist is the Euclidean distance, RBF is the radial basis function, b is the output layer bias parameter, and specifically, the radial basis function is:

6. The method according to claim 1, characterized in that The step of inferring the predicted value of the target parameter according to the actual rotation speed, the actual vibration response, the fitting relationship and the ratio comprises: According to the actual speed, solving the fitting relationship represented by the RBF proxy model by the OLS method to obtain a target ratio corresponding to the actual speed; A predicted value of the target parameter is determined based on the target ratio and the actual vibration response.

7. A rotor system response prediction system based on a digital twin model, characterized in that: The system comprises: A physical subunit, the physical subunit is used to obtain an actual rotational speed of the target rotor system and an actual vibration response at the actual rotational speed; An offline subunit, wherein the offline subunit is used to establish a dynamic model of a target rotor system by using a finite element method; A digital twin subunit, the digital twin subunit is used to determine the vibration response data of the target rotor system under different working conditions according to the dynamic model; determine the target parameters that affect the vibration response of the target rotor system according to the vibration response data, and obtain the parameter data of the target parameters under different working conditions; determine the parameter data, the rotational speed under different working conditions and the highly simplified model of the vibration response data through the GA-BPNN model; obtain the ratio of the vibration response data and the parameter data, and fit the fitting relationship between the rotational speed and the ratio through the RBF proxy model; obtain the actual rotational speed of the target rotor system and the actual vibration response at the actual rotational speed, and reversely infer the predicted value of the target parameter according to the actual rotational speed, the actual vibration response, the fitting relationship and the ratio; update the highly simplified model according to the predicted value to obtain an updated digital twin model, and complete the response prediction of the difficult-to-measure points in the rotor system based on the updated digital twin model.

8. The system according to claim 7, characterized in that The physical subunit and the digital twin subunit use the WebSocket full-duplex communication protocol to exchange data.

9. The system according to claim 7, characterized in that The digital twin subunit is a visualization unit built based on JavaScript.

10. The system according to claim 7, characterized in that The digital twin subunit is a data processing unit constructed by jointly programming MATLAB and LabVIEW based on MATLABScript.

Images