A method for dynamic evolution of digital twin models of rotor systems

Abstract

This invention discloses a method for the dynamic evolution of a digital twin model of a rotor system. First, finite element models of each component of the rotor system are established and modified to obtain a modified rotor system simulation model. Second, based on residual stiffness theory, the dynamic evolution curves of the rotor system state with varying operating conditions and service time are analyzed from three aspects: bearing wear, bolt loosening, and structural fatigue. Then, simulation and experimental data generated by the modified rotor system simulation model and the rotor system test bench under different evolution states under varying operating conditions are collected. The correlation of multivariate evolution is analyzed based on a multi-objective optimization algorithm to obtain accurate dynamic evolution curves of the rotor system state. Finally, the obtained accurate dynamic evolution curves of the rotor system state are input into the modified rotor system simulation model to obtain a dynamic evolution digital twin model of the rotor system.

Description

Technical Field

[0001] This invention relates to the field of digital twin technology, and mainly to a method for dynamic evolution of a digital twin model of a rotor system. Background Technology

[0002] With the continuous development of finite element simulation technology and the constant improvement of modal testing methods, research on digital twin modeling of mechanical equipment such as rotor systems has received widespread attention. Due to its powerful virtual sensing capabilities, digital twin models have broad application prospects in fault mechanism analysis, operational status monitoring and prediction, and intelligent equipment maintenance, and related technologies are considered an important component of intelligent manufacturing. However, the operating status of mechanical equipment such as rotor systems changes continuously with the working environment and service time. Therefore, accurately describing the operating status of rotor systems and other mechanical equipment under different working environments and service times in the digital twin model is extremely important and more in line with actual needs. To this end, this paper uses a multi-objective optimization algorithm to adjust the parameters of the digital twin model based on the difference between experimental signals of the rotor system collected under different working environments and service times, thereby endowing the digital twin model with dynamic evolution capabilities. Summary of the Invention

[0003] Purpose of the invention: Based on the problems existing in the above-mentioned background technology, the present invention provides a dynamic evolution method for a digital twin model of a rotor system, which can endow the digital twin model of a rotor system with high-precision dynamic evolution capability based on model correction technology, mechanism analysis and parameter multi-objective optimization technology.

[0004] To achieve the above objectives, the present invention adopts the following technical solution:

[0005] A method for dynamic evolution of a digital twin model of a rotor system includes the following steps:

[0006] Step S1: Establish finite element models of the eccentric disk, shaft, shaft bolt, bearing housing, bracket, bracket bolt, and bearing respectively. Obtain physical objects of the eccentric disk, shaft, shaft bolt, bearing housing, bracket, bracket bolt, and bearing respectively. Perform modal tests on the physical objects. Then, correct the established finite element models based on the modal test results. Finally, assemble the corrected finite element models to obtain the rotor system simulation model after model correction.

[0007] Step S2: Analyze the state evolution process of bearing wear, bolt loosening and structural fatigue of the rotor system as the operating conditions change and the service time increases, and use the residual stiffness theory to draw the state evolution curves of bearing wear, bolt loosening and structural fatigue.

[0008] Step S3: Based on the model obtained in Step S1, the rotor system simulation model and the rotor system test bench are respectively used to collect simulation data and experimental data under different evolution states of variable working conditions. The correlation of multivariate evolution is analyzed based on the multi-objective optimization algorithm to improve the evolution curves of bearing wear, bolt loosening and structural fatigue state obtained in Step S2, and finally obtain the accurate dynamic evolution process curve of rotor system state.

[0009] Step S4: Input the accurate rotor system state dynamic evolution process curve obtained in step S3 into the rotor system simulation model after model correction obtained in step S1 to obtain the rotor system dynamic evolution digital twin model.

[0010] Preferably, the implementation process of step S1 is as follows:

[0011] Step S1.1: Establish finite element models of the eccentric disk, shaft, shaft bolt, bearing housing, bracket, bracket bolt, and bearing respectively. Then, conduct modal tests on the actual eccentric disk, shaft, shaft bolt, bearing housing, bracket, bracket bolt, and bearing. Based on the modal test results, revise the established finite element models of the eccentric disk, shaft, shaft bolt, bearing housing, bracket, bracket bolt, and bearing. At the same time, obtain the modal frequencies, damping, and stiffness parameters of different components. The different components include the actual eccentric disk, shaft, shaft bolt, bearing housing, bracket, bracket bolt, and bearing.

[0012] The correction of the finite element model must ensure that the modal confidence criterion (MAC) calculation results for each mode of vibration of different components are accurate. ij For modal confidence scores greater than 85%, the calculation method is as follows:

[0013]

[0014] Among them, MAC ij The correlation coefficient between the i-th experimental mode shape and the j-th finite element model mode shape; The i-th experimental mode shape; Let be the j-th simulated mode shape; Τ is the conjugate transpose;

[0015] Step S1.2: Connect the modified eccentric disk finite element model and the modified shaft finite element model with thin-layer elements to form the transmission part finite element model. Modal tests are conducted on the shaft bolts to further modify the thin-layer elements of the transmission part finite element model. Connect the modified bearing seat finite element model and the modified bracket finite element model with thin-layer elements to form the support part finite element model. Modal tests are conducted on the bracket bolts to further modify the thin-layer elements of the support part finite element model.

[0016] Step S1.3: Assemble the finite element models of the transmission part, the support part, and the bearing to obtain the rotor system simulation model after model correction, and input the modal frequencies, damping, and stiffness of different components obtained from the modal tests in step S1.1 as initial parameters into the rotor system simulation model.

[0017] Preferably, the implementation process of step S2 is as follows:

[0018] Step S2.1: Assemble the eccentric disk, shaft, shaft bolts, bearing housing, bracket, bracket bolts, and bearings into a rotor system. Analyze the evolution of bearing wear as the rotor system's operating conditions change and service time increases, and plot the bearing wear state evolution curve using residual stiffness theory.

[0019]

[0020] in, The corrected basic bearing rated life at speed n; n is the bearing speed; C is the basic rated dynamic load of the bearing; η is the bearing life index; f p F is the load factor; r Radial load; This is the life correction factor; as the bearing wear condition evolves, the remaining stiffness of the bearing changes as follows:

[0021]

[0022] Among them, E b (0) represents the initial stiffness parameter of the bearing; E b (n) represents the residual stiffness due to bearing wear; θ t b This refers to the service life of the bearing.

[0023] Step S2.2: Analyze the evolution of the loosening state of the shaft bolts and support bolts as the rotor system's operating conditions change and service time increases, and use residual stiffness theory to plot the evolution curves of the loosening state of the shaft bolts and support bolts:

[0024] Measure the torque of the shaft bolts and bracket bolts at different stages, and calculate the tightening force of the shaft bolts or bracket bolts according to the following formula:

[0025] F f =T f / κd (4)

[0026] Among them, T f F is the tightening torque of the shaft bolt or bracket bolt; fκ is the tightening force of the shaft bolt or bracket bolt; d is the nominal diameter of the shaft bolt or bracket bolt; as the loosening state of the shaft bolt or bracket bolt evolves, the remaining stiffness of the connection of the shaft bolt or bracket bolt changes as follows:

[0027]

[0028] Among them, E q (0) represents the initial stiffness parameter of the connection between the shaft bolt or the bracket bolt; E q (n) represents the residual stiffness caused by loosening of shaft bolts or bracket bolts; T t The tightening torque of the shaft bolt or bracket bolt at service time t;

[0029] Step S2.3: Analyze the structural fatigue state evolution process as the rotor system's operating conditions change and service time increases, and use residual stiffness theory to plot the structural fatigue state evolution curve:

[0030] The structural fatigue life of the rotor system is calculated as follows:

[0031]

[0032] Among them, S b S represents the tensile strength of the material. ae The fatigue limit of the material; denoted by ψ, representing the shape parameter of the material; ψ represents the proportional parameter of the material; further fatigue damage of the rotor system structure is as follows:

[0033]

[0034] Where N(S) is the fatigue life of the material corresponding to stress level S; ν is the number of stress cycles per unit time; Let be the time corresponding to the stress variation interval; D be the structural fatigue damage; p(S) be the probability density function of the stress amplitude, which is solved as follows:

[0035]

[0036] Among them, variables variable Variable D3 = 1 - D1 - D2; variable variable variable variable Where m0, m1, m2, and m4 are parameters related to the working conditions; γ represents irregular factors; as the structural fatigue state evolves, the residual stiffness of the structure changes as follows:

[0037] B(n) = B(0) - (B(0) - B ND (9)

[0038] Where B(n) is the residual stiffness caused by structural fatigue; B(0) is the initial stiffness of the rotor system; B N This refers to the fatigue stiffness loss of the rotor system structure.

[0039] Preferably, the implementation process of step S3 is as follows:

[0040] Step S3.1: Assemble the acquired eccentric disk, shaft, shaft bolts, bearing housing, bracket, bracket bolts, and bearings into a rotor system test bench. Collect experimental data from the rotor system test bench under different evolutionary states of varying operating conditions, as shown below: Where H represents the number of operating conditions; G represents the number of evolution states; further, simulation data were collected based on the modified rotor system simulation model under the same operating conditions as in the experiment, and are expressed as follows:

[0041] Step S3.2: Based on the multi-objective optimization algorithm, describe the correlation of multivariate evolution, and finally obtain the accurate dynamic evolution process curve of the rotor system state. The objective function of the multi-objective optimization algorithm is designed as follows:

[0042]

[0043] Where J1 and J2 are the objective functions in the time domain and frequency domain, respectively; It is the frequency; FFT(·) is the Fourier transform function; by optimizing the residual stiffness curves under different evolution states of the variable working conditions obtained in steps S2.1, S2.2 and S2.3, i.e. formulas (3)(5)(9), the objective function (10) is minimized, and finally the accurate dynamic evolution process curve of the rotor system state is obtained.

[0044] Preferably, the implementation process of step S4 is as follows:

[0045] The accurate bearing wear state evolution curve, bolt loosening state evolution curve, and structural fatigue state evolution curve obtained in step S3.2 are input into the modified rotor system simulation model, and finally the dynamic evolution digital twin model of the rotor system is obtained.

[0046] Beneficial effects:

[0047] (1) The established digital twin model of the rotor system has dynamic evolution capability, which can accurately describe the state changes of the rotor system under different working conditions and service time, and thus provide a reliable model reference for actual engineering needs.

[0048] (2) The established digital twin model of the rotor system has significant advantages over traditional test benches and measurement technologies. It can extract and visualize signals of any position and any type at a lower cost. It has great application value in fault mechanism analysis, operation status monitoring and prediction and intelligent equipment maintenance.

[0049] (2) Based on the mechanism analysis system, the parameter evolution process of bearing wear, bolt loosening and structural fatigue in the rotor system under different service times under different working conditions was studied, and the residual stiffness theory was used to characterize the evolution state.

[0050] (3) Based on the multi-objective optimization algorithm, the correlation of multi-variable state evolution such as bearing wear, bolt loosening and structural fatigue was analyzed, and the accurate dynamic evolution process curve of the rotor system was obtained. Attached Figure Description

[0051] Figure 1 A flowchart of a dynamic evolution method for a digital twin model of a rotor system provided by the present invention;

[0052] Figure 2 The calculation results of the modified modal confidence criterion for the rotor system support model in this embodiment of the invention;

[0053] Figure 3 The calculation results of the modified modal confidence criterion for the rotor system bearing housing model in this embodiment of the invention;

[0054] Figure 4 This refers to the thin-layer unit for shaft bolts established in this embodiment of the invention;

[0055] Figure 5 This refers to the thin-layer unit of the support bolt established in the embodiments of the present invention;

[0056] Figure 6 This is the simulation model structure of the rotor system after model correction in this embodiment of the invention;

[0057] Figure 7 This is the modified basic rated life curve of the bearing in this embodiment of the invention;

[0058] Figure 8 This represents the initial bolt state in an embodiment of the present invention.

[0059] Figure 9 This refers to the state of the intermediate bolt in an embodiment of the present invention;

[0060] Figure 10 This represents the final bolt state in this embodiment of the invention.

[0061] Figure 11 This is the bolt loosening state evolution curve in an embodiment of the present invention;

[0062] Figure 12 The stress analysis results of the finite element model of the rotor system in this embodiment of the invention;

[0063] Figure 13 This refers to the stress response power spectral density of the rotor system under varying operating conditions in this embodiment of the invention.

[0064] Figure 14 The fatigue life curve of the rotor system structure in this embodiment of the invention;

[0065] Figure 15 This is the structural fatigue state evolution curve in an embodiment of the present invention;

[0066] Figure 16 This is a flowchart illustrating the analysis of multivariate evolutionary correlations based on a multi-objective optimization algorithm in an embodiment of the present invention;

[0067] Figure 17 This is a bearing wear state evolution curve considering multivariate correlations in an embodiment of the present invention;

[0068] Figure 18 This is a bolt loosening state evolution curve considering multivariate correlation in an embodiment of the present invention;

[0069] Figure 19 This is a structural fatigue state evolution curve considering multivariate correlations in an embodiment of the present invention;

[0070] Figure 20 The data represents simulation data collected from a digital twin model of the rotor system under varying operating conditions and service times, as described in this embodiment of the invention. Detailed Implementation

[0071] The invention will now be further described with reference to the accompanying drawings.

[0072] See Figures 1-20 This embodiment provides a method for the dynamic evolution of a digital twin model of a rotor system. The process of this method is as follows: Figure 1 As shown, this embodiment uses a certain type of rotor system test bench as an example to verify the algorithm performance. The strategy specifically includes the following steps:

[0073] Step S1: Establish finite element models of the transmission part consisting of the eccentric disk, shaft and shaft bolts of the rotor system, establish finite element models of the support part consisting of the bearing housing, bracket and bracket bolts, establish a bearing finite element model, further correct the established finite element models based on the modal test results and assemble them to obtain the rotor system simulation model after model correction.

[0074] Specifically, in this embodiment, step S1 includes:

[0075] Step S1.1: Establish finite element models for components of the rotor system, including the eccentric disk, shaft, bearing housing, support, and bearings. Further perform modal tests on these components to revise the established finite element models and obtain the modal frequencies, damping, and stiffness parameters of different components. Model revision must ensure that the modal confidence criterion (MAC) calculation results for each modal mode of different components are accurate. ij For modal confidence scores greater than 85%, the calculation method is as follows:

[0076]

[0077] Among them, MAC ij The correlation coefficient between the i-th experimental mode shape and the j-th finite element model mode shape; The i-th experimental mode shape; Γ represents the j-th simulated mode shape; Τ represents the conjugate transpose. Figure 2 and Figure 3 The calculation results of the modified modal confidence criteria for the system support and rotor system bearing housing models are presented respectively. It can be found that the MAC ij All values ​​are greater than 85%, indicating that the model correction accuracy meets the requirements. Table 1 shows the model correction results for each component of the rotor system.

[0078] Table 1. Model correction results for each component of the rotor system

[0079]

[0080] Step S1.2: Connect the eccentric disk finite element model corrected in Step S1.1 and the shaft finite element model corrected in Step S1.1 respectively using thin-layer elements to form the transmission part finite element model. The thin-layer elements of the transmission part finite element model are corrected by testing the shaft bolt modes. The shaft bolt thin-layer elements are as follows: Figure 4 As shown; the finite element model of the bearing housing modified in step S1.2 and the finite element model of the bracket modified in step S1.2 are respectively connected by thin-layer elements to form the finite element model of the support part. The thin-layer elements of the support part finite element model are modified by testing the bracket bolt modes. The thin-layer elements of the bracket bolt are as follows. Figure 5 As shown in Table 2, the model correction results for the finite element models of the transmission and support parts are presented.

[0081] Table 2. Model Correction Results of Finite Element Models for Transmission and Support Parts

[0082]

[0083] Step S1.3: Assemble the modified finite element model of the transmission part obtained in step S1.2, the modified finite element model of the support part obtained in step S1.2, and the modified finite element model of the bearing obtained in step S1.1 to obtain the modified rotor system simulation model. Input the modal frequencies, damping, and stiffness of different components of the rotor system obtained from the modified model into the rotor system simulation model as initial parameters. The final structure of the modified rotor system simulation model is as follows: Figure 6 As shown.

[0084] Step S2: Analyze the state evolution process of bearing wear, bolt loosening and structural fatigue of the rotor system as the operating conditions change and the service time increases, and use the residual stiffness theory to draw the state evolution curves of bearing wear, bolt loosening and structural fatigue.

[0085] Specifically, in this embodiment, step S2 includes:

[0086] Step S2.1: Analyze the evolution of bearing wear as the rotor system's operating conditions change and service time increases, and plot the bearing wear state evolution curve using residual stiffness theory:

[0087]

[0088] in, The bearing's basic rated life at the corrected rotational speed n; n is the bearing rotational speed; C is the bearing's basic rated dynamic load; η is the bearing life index; f p F is the load factor; r Radial load; This is the life correction factor; as the bearing wear condition evolves, the remaining stiffness of the bearing changes as follows:

[0089]

[0090] Among them, E b (0) represents the initial bearing stiffness obtained in step S1.3; E b (n) represents the residual stiffness due to bearing wear; θ t b The bearing service life is shown in Table 3.

[0091] Table 3 Bearing structural parameters

[0092] parameter value parameter value Outer radius 40mm Basic rated dynamic load 62Kn Inner radius 20mm Rated speed 9500r / min Load factor 1.2 Bearing life index 10 / 3

[0093] The calculated basic rated lives of the bearings at speeds of 1000 r / min, 2000 r / min, and 3000 r / min are as follows:

[0094]

[0095] The modified bearing basic rated life curve obtained according to equation (4) is as follows: Figure 7 As shown.

[0096] Step S2.2: Analyze the evolution of bolt loosening as the rotor system's operating conditions change and service time increases, and plot the bolt loosening state evolution curve using residual stiffness theory:

[0097] Measure the bolt torque at different stages and calculate the bolt tightening force according to the following formula:

[0098] F f =T f / κd (5)

[0099] Among them, T f F represents the bolt tightening torque. f κ is the bolt tightening force; d is the nominal diameter of the thread; as the bolt loosening state evolves, the remaining stiffness of the bolt connection changes as follows:

[0100]

[0101] Among them, E q (0) represents the initial stiffness of the bolted connection obtained in step S1.3; E q (n) represents the residual stiffness caused by bolt loosening; T t The bolt tightening torque at service time t; simulation models of bolt states under different degrees of loosening, such as... Figures 8-10 As shown in the figure. The final obtained bolt loosening state evolution curve is as follows. Figure 11 As shown.

[0102] Step S2.3: Analyze the structural fatigue state evolution process as the rotor system's operating conditions change and service time increases, and plot the structural fatigue state evolution curve using residual stiffness theory: The stress analysis results of the rotor system finite element model are as follows... Figure 12 As shown; the stress response power spectral density of the rotor system under varying operating conditions is as follows: Figure 13 As shown.

[0103] The fatigue life of the rotor system structure is calculated as follows:

[0104]

[0105] Among them, S b S represents the tensile strength of the material. ae The fatigue limit of the material; ψ represents the shape parameter of the material; ψ represents the proportional parameter of the material. The fatigue life curve of the rotor system structure is obtained as follows: Figure 14 As shown in Table 4, the material parameters of the rotor system are as follows:

[0106] Table 4 Material parameters of the rotor system

[0107]

[0108]

[0109] Further characterization of fatigue damage in the rotor system structure is as follows: [The following is a list of results, not a direct translation]

[0110]

[0111] Where N(S) is the fatigue life of the material corresponding to stress level S; ν is the number of stress cycles per unit time; Let be the time corresponding to the stress variation interval; D be the structural fatigue damage; p(S) be the probability density function of the stress amplitude, which needs to be solved as follows:

[0112]

[0113] in, D3 = 1 - D1 - D2; Where m0, m1, m2 and m4 are parameters related to the working conditions; γ represents irregular factors; the parameter values ​​of the probability density function model of stress amplitude are shown in Table 5.

[0114] Table 5. Parameter values ​​of the probability density function model for stress amplitude.

[0115] Speed ​​(r / min) <![CDATA[m0]]> <![CDATA[m1]]> <![CDATA[m2]]> <![CDATA[m4]]> 1000 93.22 <![CDATA[2.43×10 4 ]]> <![CDATA[7.97×10 6 ]]> <![CDATA[1.52×10 12 ]]> 2000 163.22 <![CDATA[4.24×10 4 ]]> <![CDATA[1.39×10 7 ]]> <![CDATA[2.76×10 12 ]]> 3000 261.19 <![CDATA[6.63×10 4 ]]> <![CDATA[2.06×10 7 ]]> <![CDATA[3.61×10 12 ]]>

[0116] As the structural fatigue state evolves, the residual stiffness of the structure changes as follows:

[0117] B(n) = B(0) - (B(0) - B N D (10)

[0118] Where B(n) is the residual stiffness caused by structural fatigue; B(0) is the initial stiffness of the rotor system structure obtained in step S1.3; B N The stiffness loss of the rotor system structure; the final structural fatigue state evolution curve is as follows: Figure 15 As shown.

[0119] Step S3: Based on the model obtained in Step S1, the rotor system simulation model and the rotor system test bench are modified and simulation data and experimental data are collected under different evolution states of variable operating conditions. The correlation of multivariate evolution is analyzed based on a multi-objective optimization algorithm to improve the bearing wear, bolt loosening, and structural fatigue state evolution curves obtained in Step S2, ultimately obtaining accurate dynamic evolution process curves of the rotor system state; the process is as follows: Figure 16 As shown.

[0120] Specifically, in this embodiment, step S3 includes:

[0121] Step S3.1: Collect experimental data from the rotor system test bench under different evolution states of variable operating conditions, as shown below: Where H represents the number of operating conditions; G represents the number of evolution states; further, simulation data were collected based on the modified rotor system simulation model under the same operating conditions as in the experiment, and are expressed as follows:

[0122] Step S3.2: Based on the multi-objective optimization algorithm, describe the correlation of multivariate evolution, and finally obtain the accurate dynamic evolution process curve of the rotor system state. The objective function of the multi-objective optimization algorithm is designed as follows:

[0123]

[0124] Where J1 and J2 are the objective functions in the time domain and frequency domain, respectively; It is the frequency; FFT(·) is the Fourier transform function; by optimizing the residual stiffness curves under different evolution states of the variable working conditions obtained in steps S2.1, S2.2 and S2.3, the objective function (11) is minimized; minimizing the objective function (11) means that the experimental data collected based on the rotor system test bench and the simulation data collected based on the modified rotor system simulation model are highly consistent in the frequency domain and time domain, and finally the accurate rotor system state dynamic evolution process curve is obtained; Figures 17-19 These are the evolution curves of bearing wear, bolt loosening, and structural fatigue state considering multivariate correlations.

[0125] Step S4: Input the accurate rotor system state dynamic evolution process curve obtained in Step S3 into the rotor system simulation model after model correction obtained in Step S1 to obtain the rotor system dynamic evolution digital twin model:

[0126] Specifically, in this embodiment, step S4 includes:

[0127] The accurate bearing wear state evolution curve, bolt loosening state evolution curve, and structural fatigue state evolution curve obtained in step S3.2 are input into the rotor system simulation model after model correction obtained in step S1.3. That is, the stiffness parameters of the thin-layer elements corresponding to the bearings and bolts and the structural finite element model in the rotor system finite element model evolve continuously with the change of working conditions and the increase of service time; finally, a dynamic evolution digital twin model of the rotor system is obtained. Figure 20 This is simulation data collected based on a digital twin model of the rotor system under varying operating conditions and service times.

[0128] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for dynamic evolution of a digital twin model of a rotor system, characterized in that, Includes the following steps: Step S1: Establish finite element models of the eccentric disk, shaft, shaft bolt, bearing housing, bracket, bracket bolt, and bearing respectively. Obtain physical objects of the eccentric disk, shaft, shaft bolt, bearing housing, bracket, bracket bolt, and bearing respectively. Perform modal tests on the physical objects. Then, correct the established finite element models based on the modal test results. Finally, assemble the corrected finite element models to obtain the rotor system simulation model after model correction. Step S2: Analyze the state evolution process of the rotor system as the operating conditions change and service time increases, including bearing wear, bolt loosening, and structural fatigue, and plot the state evolution curves of bearing wear, bolt loosening, and structural fatigue using residual stiffness theory; the specific implementation process is as follows: Step S2.1: Assemble the eccentric disk, shaft, shaft bolts, bearing housing, bracket, bracket bolts, and bearings into a rotor system. Analyze the evolution of bearing wear as the rotor system's operating conditions change and service time increases, and plot the bearing wear state evolution curve using residual stiffness theory. in, The corrected basic bearing rated life at speed n; n is the bearing speed; C is the basic rated dynamic load of the bearing; η is the bearing life index; f p F is the load factor; r Radial load; This is the life correction factor; as the bearing wear condition evolves, the remaining stiffness of the bearing changes as follows: Among them, E b (0) represents the initial stiffness parameter of the bearing; E b (n) represents the residual stiffness due to bearing wear; This refers to the service life of the bearing. Step S2.2: Analyze the evolution of the loosening state of the shaft bolts and support bolts as the rotor system's operating conditions change and service time increases, and use residual stiffness theory to plot the evolution curves of the loosening state of the shaft bolts and support bolts: Measure the torque of the shaft bolts and bracket bolts at different stages, and calculate the tightening force of the shaft bolts or bracket bolts according to the following formula: F f < T f / κd (4) Among them, T f F is the tightening torque of the shaft bolt or bracket bolt; f κ is the tightening force of the shaft bolt or bracket bolt; d is the nominal diameter of the shaft bolt or bracket bolt; as the loosening state of the shaft bolt or bracket bolt evolves, the remaining stiffness of the connection of the shaft bolt or bracket bolt changes as follows: Among them, E q (0) represents the initial stiffness parameter of the connection between the shaft bolt or the bracket bolt; E q (n) represents the residual stiffness caused by loosening of shaft bolts or bracket bolts; T t The tightening torque of the shaft bolt or bracket bolt at service time t; Step S2.3: Analyze the structural fatigue state evolution process as the rotor system's operating conditions change and service time increases, and use residual stiffness theory to plot the structural fatigue state evolution curve: The structural fatigue life of the rotor system is calculated as follows: Among them, S b S represents the tensile strength of the material. ae The fatigue limit of the material; denoted by ψ, representing the shape parameter of the material; ψ represents the proportional parameter of the material; further fatigue damage of the rotor system structure is as follows: Where N(S) is the fatigue life of the material corresponding to stress level S; ν is the number of stress cycles per unit time; θ is the time corresponding to the stress variation interval; D is the structural fatigue damage; and p(S) is the probability density function of stress amplitude, which is solved as follows: Among them, variables variable Variable D3 = 1 - D1 - D2; variable variable variable variable Where m0, m1, m2, and m4 are parameters related to the working conditions; γ represents irregular factors; as the structural fatigue state evolves, the residual stiffness of the structure changes as follows: B(n)=B(0)-(B(0)-B N )D (9) Where B(n) is the residual stiffness caused by structural fatigue; B(0) is the initial stiffness of the rotor system; B N This refers to the fatigue stiffness loss of the rotor system structure; Step S3: Based on the model obtained in Step S1, the rotor system simulation model and the rotor system test bench are respectively used to collect simulation data and experimental data under different evolution states of variable working conditions. The correlation of multivariate evolution is analyzed based on the multi-objective optimization algorithm to improve the evolution curves of bearing wear, bolt loosening and structural fatigue state obtained in Step S2, and finally obtain the accurate dynamic evolution process curve of rotor system state. Step S4: Input the accurate rotor system state dynamic evolution process curve obtained in step S3 into the rotor system simulation model after model correction obtained in step S1 to obtain the rotor system dynamic evolution digital twin model.

2. The dynamic evolution method for a digital twin model of a rotor system according to claim 1, characterized in that, The implementation process of step S1 is as follows: Step S1.1: Establish finite element models of the eccentric disk, shaft, shaft bolt, bearing housing, bracket, bracket bolt, and bearing respectively. Then, conduct modal tests on the actual eccentric disk, shaft, shaft bolt, bearing housing, bracket, bracket bolt, and bearing. Based on the modal test results, revise the established finite element models of the eccentric disk, shaft, shaft bolt, bearing housing, bracket, bracket bolt, and bearing. At the same time, obtain the modal frequencies, damping, and stiffness parameters of different components. The different components include the actual eccentric disk, shaft, shaft bolt, bearing housing, bracket, bracket bolt, and bearing. The correction of the finite element model must ensure that the modal confidence criterion (MAC) calculation results for each mode of vibration of different components are accurate. ij For modal confidence scores greater than 85%, the calculation method is as follows: Among them, MAC ij The correlation coefficient between the i-th experimental mode shape and the j-th finite element model mode shape; The i-th experimental mode shape; Let be the j-th simulated mode shape; Τ is the conjugate transpose; Step S1.2: Connect the modified eccentric disk finite element model and the modified shaft finite element model with thin-layer elements to form the transmission part finite element model. Modal tests are conducted on the shaft bolts to further modify the thin-layer elements of the transmission part finite element model. Connect the modified bearing seat finite element model and the modified bracket finite element model with thin-layer elements to form the support part finite element model. Modal tests are conducted on the bracket bolts to further modify the thin-layer elements of the support part finite element model. Step S1.3: Assemble the finite element models of the transmission part, the support part, and the bearing to obtain the rotor system simulation model after model correction, and input the modal frequencies, damping, and stiffness of different components obtained from the modal tests in step S1.1 as initial parameters into the rotor system simulation model.

3. The method for dynamic evolution of a digital twin model of a rotor system according to claim 1, characterized in that, The implementation process of step S3 is as follows: Step S3.1: Assemble the acquired eccentric disk, shaft, shaft bolts, bearing housing, bracket, bracket bolts, and bearings into a rotor system test bench. Collect experimental data from the rotor system test bench under different evolutionary states of varying operating conditions, as shown below: Where H represents the number of operating conditions; G represents the number of evolution states; further, simulation data were collected based on the modified rotor system simulation model under the same operating conditions as in the experiment, and are expressed as follows: Step S3.2: Based on the multi-objective optimization algorithm, describe the correlation of multivariate evolution, and finally obtain the accurate dynamic evolution process curve of the rotor system state. The objective function of the multi-objective optimization algorithm is designed as follows: Where J1 and J2 are the objective functions in the time domain and frequency domain, respectively; It is the frequency; FFT(g) is the Fourier transform function; by optimizing the residual stiffness curves under different evolution states of variable working conditions obtained in steps S2.1, S2.2 and S2.3, i.e. formulas (3)(5)(9), the objective function (10) is minimized, and finally the accurate dynamic evolution process curve of the rotor system state is obtained.

4. The dynamic evolution method for a digital twin model of a rotor system according to claim 3, characterized in that, The implementation process of step S4 is as follows: The accurate bearing wear state evolution curve, bolt loosening state evolution curve, and structural fatigue state evolution curve obtained in step S3.2 are input into the modified rotor system simulation model, and finally the dynamic evolution digital twin model of the rotor system is obtained.

Images