Mechanical component reliability digital twin model construction method and computer equipment

Abstract

The present invention discloses a method for constructing a reliability digital twin model of a mechanical component and a computer device. The method constructs reliability digital twin models of mechanical components in sequence according to the priority order of a digital twin model of a mechanical component based on a failure mechanism model, a digital twin model of a mechanical component based on an empirical model, a digital twin model of a mechanical component based on a simulation model, a digital twin model of a mechanical component based on a structure function function, and a digital twin model of a mechanical component based on a data model. Thus, a complete mechanical component reliability digital twin model can be constructed, and simulation, monitoring, diagnosis, prediction and control of reliability indicators such as failure rate, reliability, life and remaining life of the mechanical component can be achieved. Reliability design analysis, test evaluation and diagnosis prediction of mechanical components can be achieved by digital means, thereby greatly shortening the product iteration and upgrade cycle and greatly reducing the research and development and use costs of mechanical components.

Description

Technical Field

[0001] The present invention belongs to the technical field of reliability and relates to a method for constructing a digital twin model of mechanical component reliability and a computer device. Background Art

[0002] Mechanical parts are important components of mechanical products. The use environment of mechanical products is complex and changeable, and the existing test environment is often difficult to fully simulate the actual use conditions, resulting in deviations between the test results and the actual use conditions. The problem of insufficient test environment simulation is prominent. In addition, due to the wide variety of mechanical products, the parts vary greatly, and the test cycle is long and the cost is high, resulting in insufficient test samples. It is difficult to obtain accurate reliability data, which can no longer meet the current needs of rapid product iteration, quality improvement and efficiency increase. As an emerging simulation technology, reliability digital twin technology realizes real-time simulation, monitoring, prediction and optimization of the reliability operation status of physical objects by creating digital mirrors of physical objects. With the rapid development of computer technology and the widespread application of technologies such as big data and artificial intelligence, reliability digital twin technology has gradually expanded to multiple fields. In the reliability design analysis, test evaluation and operation monitoring of mechanical products, reliability digital twin technology has provided a powerful solution for improving the reliability of mechanical products with its unique advantages.

[0003] At present, the application of digital twin technology is mainly focused on function and performance, but there is no digital twin model of mechanical component reliability. It is impossible to simulate, monitor, diagnose, predict and control the reliability characteristics, reliability behavior and reliability process of mechanical components in the digital space, and it is difficult to meet the needs of forward design, rapid testing and diagnostic prediction of mechanical component reliability. Summary of the invention

[0004] The purpose of the present invention is to provide a method for constructing a digital twin model of mechanical component reliability, a computer device, a computer-readable storage medium and a computer program product, which can construct a complete digital twin model of mechanical component reliability and realize the reliability design analysis, test evaluation and diagnostic prediction of mechanical components by digital means.

[0005] In order to achieve the above object, one aspect of the present invention provides a method for constructing a mechanical component reliability digital twin model, comprising:

[0006] Step S1: for the i-th mechanical component, determine whether there is a failure mechanism model. If the determination is no, execute step S2. If the determination is yes, construct a digital twin model of the mechanical component based on the failure mechanism model:

[0007]

[0008] Among them, RDT1(i) represents the reliability digital twin model of the i-th mechanical component based on the failure mechanism model, It represents the mechanism model that complies with the FMI protocol formed by the failure mechanism model of the i-th mechanical component through simulation;

[0009] Step S2, determine whether there is an empirical model. If the judgment is no, execute step S3. If the judgment is yes, construct a digital twin model of the mechanical component based on the empirical model:

[0010]

[0011] Among them, RDT2 (i) represents the reliability digital twin model of the i-th mechanical component based on the empirical model, The empirical model of the i-th mechanical component is formed by simulation and conforms to the FMI protocol;

[0012] Step S3, determine whether simulation is possible, if it is not, execute step S4, if it is yes, construct a digital twin model of the mechanical component based on the simulation model:

[0013]

[0014] Among them, RDT3 (i) represents the reliability digital twin model of the i-th mechanical component based on the simulation model, The simulation model of the i-th mechanical component is formed by simulation and conforms to the FMI protocol;

[0015] Step S4, determine whether there is a structure-function function. If it is determined to be no, execute step S5. If it is determined to be yes, construct a digital twin model of the mechanical component based on the structure-function function model:

[0016]

[0017] Among them, RDT4 (i) represents the reliability digital twin model of the i-th mechanical component based on the structure-function function model, The structure function function model of the i-th mechanical component is formed by simulation and conforms to the FMI protocol;

[0018] Step S5, determine whether there is historical data. If yes, build a digital twin model of the mechanical component based on the data model:

[0019]

[0020] Among them, RDT5 (i)represents the reliability digital twin model of the i-th mechanical component based on the data model, The data model representing the i-th mechanical component is formed through simulation and is in compliance with the FMI protocol;

[0021] Step S6, if the judgment in step S5 is no, obtain the test data as historical data through test and repeat step S5, and then judge whether the modeling of all mechanical components is completed. If the judgment is no, return to step S1 to model the next mechanical component until the reliability modeling of all mechanical components is completed.

[0022] Another aspect of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above method.

[0023] Another aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above method when executed by a processor.

[0024] Another aspect of the present invention provides a computer program product, comprising a computer program, which implements the steps of the above method when executed by a processor.

[0025] According to the mechanical component reliability digital twin model construction method, computer equipment, computer-readable storage medium and computer program product of the above aspects of the present invention, a complete mechanical component reliability digital twin model can be constructed, and reliability design analysis, test evaluation and diagnostic prediction of mechanical components can be realized by digital means. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings used in the description of the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work:

[0027] Figure 1 It is an overall flow chart of a method for constructing a mechanical component reliability digital twin model according to an embodiment of the present invention;

[0028] Figure 2 It is a schematic diagram of the construction of each model in the method for constructing a digital twin model of mechanical component reliability according to an embodiment of the present invention;

[0029] Figure 3 4 is a structural diagram of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. 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.

[0031] The embodiments of the present invention focus on reliability and provide a method for constructing a digital twin model of mechanical component reliability, thereby providing a digital model basis for mechanical component reliability design analysis, test evaluation, and diagnostic prediction.

[0032] A mechanical system is often composed of multiple mechanical components. To build a reliability digital twin of the entire mechanical system, it is necessary to build digital twins of all the mechanical components included separately, and then based on the digital twins of these mechanical components, finally form a reliability digital twin of the entire mechanical system.

[0033] According to the source of mechanical component failure models, five modeling methods are covered:

[0034] First, build a digital twin model of mechanical components based on failure mechanism models. The failure mechanisms include wear, aging, corrosion, fatigue and other loss-type mechanisms, which are mainly derived from mechanical standards, mechanical manuals, and models established by mature mechanism models in the mechanical industry. Establish a mapping relationship between the input parameters of mechanical components and reliability indicators such as life, failure rate, and reliability.

[0035] Second, construct a digital twin model of mechanical components based on an empirical model. This empirical model is mainly derived from the empirical formula model established by engineers’ experience and industry experience, and establishes a mapping relationship between the input parameters of mechanical components and reliability indicators such as life, failure rate, and reliability.

[0036] Third, a digital twin model of mechanical components with fatigue loss failure type based on the simulation model is constructed. The fatigue loss failure model is derived from the model established by the fatigue simulation software, and the mapping relationship between the input parameters of the mechanical components and reliability indicators such as life, failure rate, and reliability is established;

[0037] Fourth, construct a digital twin model of mechanical components based on the structure-function function. The structure-function function is mainly established through the limit state equations of mechanics, thermal and life, and the mapping relationship between the input parameters of the mechanical components and reliability indicators such as life, failure rate, and reliability is established;

[0038] Fifth, build a digital twin model of mechanical components based on a data model. The data model is mainly built through historical data or data obtained through experimental tests to establish a mapping relationship between the input parameters of mechanical components and reliability indicators such as life, failure rate, and reliability;

[0039] Generally, a mechanical system cannot be modeled by only one of the above methods. It is necessary to comprehensively select the above methods according to the actual situation to jointly build all the mechanical component models. Figure 1 The flowchart shows the priority logical relationship of various model selections. For the i-th mechanical component, first determine whether there is a mechanism model. If so, perform mechanism modeling. Otherwise, determine whether there is an empirical model. If so, perform empirical modeling. Otherwise, determine whether it can be simulated. If so, perform simulation-based modeling. Otherwise, determine whether there is a functional function. If so, perform functional function modeling. Otherwise, determine whether there is historical data. If so, perform data-based modeling. Otherwise, obtain test data through experimental testing and perform data-based modeling. Determine whether all mechanical component modeling is completed. If so, complete mechanical component reliability modeling. Otherwise, proceed to the judgment process for the next mechanical component. Refer to the following Figure 2 The five modeling methods of the embodiments of the present invention are described in detail in the schematic diagram.

[0040] The first type: Construction of digital twin model of mechanical components based on failure mechanism model

[0041] The premise of building the first type of mechanical component digital twin model based on the failure mechanism model is to first build its failure mechanism model. The failure mechanism includes wear, aging, corrosion, fatigue and other loss-type mechanisms, which are mainly derived from mechanical standards, mechanical manuals, and mature mechanism models in the mechanical industry. The model construction steps are as follows:

[0042] Construct the geometric model parameters of the mechanical parts to describe the size parameters and assembly parameters of the mechanical parts. The geometric model parameters of the i-th mechanical part can be expressed as

[0043] G P(i) =G1(D (i) , A (i) )

[0044] Among them, G P(i) represents the geometric model parameters of the i-th mechanical component, D (i) represents the size parameter of the i-th mechanical component, A (i) represents the assembly parameters of the i-th mechanical component.

[0045] Construct the operating state parameters of the mechanical components to describe the speed, torque, flow, current, voltage and other parameters of the mechanical components. The operating state parameters of the i-th mechanical component can be expressed as

[0046] O P(i) =O(R P(i) , T P(i) , F P(i) , C P(i) , V P(i) )

[0047] Among them, O P(i) represents the operating state parameter of the i-th mechanical component, R P(i) represents the speed parameter of the i-th mechanical component, T P(i) represents the torque parameter of the i-th mechanical component, F P(i) represents the flow parameter of the i-th mechanical component, C P(i) represents the current parameter of the i-th mechanical component, V P(i) Represents the voltage parameter of the i-th mechanical component.

[0048] Construct the external load parameters of mechanical components to describe the force load, heat load, humidity load, etc. of mechanical components. The external load parameters of the i-th mechanical component can be expressed as

[0049] P P(i) =P(L M(i) , L T(i) , L H(i) )

[0050] Among them, P P(i) represents the external load parameter of the i-th mechanical component, L M(i) represents the force load parameter of the i-th mechanical component, L T(i) represents the thermal load parameter of the i-th mechanical component, L H(i) represents the humidity load parameter of the i-th mechanical component.

[0051] Construct the structural characteristic parameters of mechanical parts to describe the material parameters, process parameters, motion parameters, electrical parameters, constant coefficients, etc. of mechanical parts. The structural characteristic parameters of the i-th mechanical part can be expressed as S P(i) =S(M C(i) , P C(i) , K C(i) , E C(i) , C C(i) )

[0052] Among them, S P(i) represents the structural characteristic parameter of the i-th mechanical component, M C(i) represents the material coefficient of the i-th mechanical component, P C(i) Represents the process coefficient of the i-th mechanical component, K C(i) represents the motion coefficient of the ith mechanical component, E C(i) represents the electrical stress coefficient of the i-th mechanical component, C C(i)represents the constant coefficient of the i-th mechanical component.

[0053] Based on the above four types of parameters, the failure mechanism model of the i-th mechanical component can be expressed as

[0054] FM model(i) (L1 (i) ,λ1 (i) , R1 (i) )=M1(G P(i) , O P(i) , P P(i) , S P(i) )

[0055] Among them, FM model(i) (L1 (i) ,λ1 (i) , R1 (i) ) represents the failure mechanism model of the i-th mechanical component, L1 (i) represents the life parameter of the output of the i-th mechanical component, λ1 (i) Represents the failure rate parameter of the i-th mechanical component output, R1 (i) Represents the reliability parameter of the output of the i-th mechanical component.

[0056] After building the failure mechanism model of the mechanical components, it is necessary to build a reliability digital twin model of the mechanical components. The steps are as follows:

[0057] Using the Simulink module in Matlab software, the mathematical expression of the failure mechanism model of the i-th mechanical component is obtained to create a simulation model in the Simulink environment, configure the step mode and step value of the model solution, and export the i-th mechanical component functional model unit FMU1 that meets the FMI (Functional Mock-up Interface) standard. (i) ,Right now

[0058]

[0059] Among them, FMU1 (i) represents the functional model unit based on the failure mechanism model of the i-th mechanical component, It represents the mechanism model that complies with the FMI protocol after simulation of the failure mechanism model of the i-th mechanical component.

[0060] Since this FMU1 (i) It has fast computing capability and a universal FMI interface for system simulation, so it can be used as a reliability digital twin model based on the failure mechanism model of the i-th mechanical component.

[0061] RDT1 (i) =FMU1(i)

[0062] Among them, RDT1 (i) Represents the reliability digital twin model of the i-th mechanical component based on the failure mechanism model.

[0063] The second type: Construction of digital twin models of mechanical components based on empirical models

[0064] The second type of mechanical component digital twin model based on empirical model is based on the premise of building its empirical model first. The empirical model is mainly derived from the empirical formula model established by engineers’ experience, industry experience, etc. The steps are as follows:

[0065] The formula of the empirical model is similar in form to the failure mechanism model and is expressed as follows:

[0066] E model(i) (L2 (i) ,λ2 (i) , R2 (i) )=E(G P(i) , O P(i) , P P(i) , S P(i) )

[0067] Among them, E model(i) (L2 (i) ,λ2 (i) , R2 (i) ) represents the empirical model of the i-th mechanical component, L2 (i) represents the life parameter of the output of the i-th mechanical component, λ2 (i) represents the failure rate parameter of the output of the i-th mechanical component, R2 (i) Represents the reliability parameter of the output of the i-th mechanical component.

[0068] After building the empirical model of the mechanical component, it is necessary to build a reliability digital twin model of the mechanical component. The construction method is similar to the digital twin model of the mechanical component based on the failure mechanism model. The steps are as follows:

[0069] Using the Simulink module in Matlab software, the mathematical expression of the empirical model of the i-th mechanical component is used to create a simulation model in the Simulink environment, the step mode and step value of the model solution are configured, and the i-th mechanical component functional model unit FMU2 that meets the FMI standard is exported. (i) ,Right now

[0070]

[0071] Among them, FMU2 (i) represents the functional model unit based on the empirical model of the i-th mechanical component, It represents the empirical model of the i-th mechanical component which is formed through simulation and conforms to the FMI protocol.

[0072] Since this FMU2 (i) It has fast computing capability and a universal FMI interface for system simulation, so it can be used as a reliability digital twin model based on an empirical model for the i-th mechanical component.

[0073] RDT2 (i) =FMU2 (i)

[0074] Among them, RDT2 (i) Represents the reliability digital twin model of the i-th mechanical component based on the empirical model.

[0075] The third category: Construction of digital twin models of mechanical components based on simulation

[0076] The third type of mechanical component digital twin model based on simulation is based on the premise of building its fatigue simulation model first. The fatigue simulation model comes from the fatigue simulation software model. The steps are as follows:

[0077] Construct a geometric model of mechanical parts to describe the size parameters and assembly parameters of mechanical parts. The geometric model parameters of the i-th mechanical part can be expressed as

[0078] G model(i) =G2(D (i) , A (i) )

[0079] Among them, G model(i) represents the geometric model of the i-th mechanical component, D (i) represents the size parameter of the i-th mechanical component, A (i) represents the assembly parameters of the i-th mechanical component.

[0080] The multi-physics simulation model of mechanical components is constructed, which mainly includes mechanical models and thermal models, which are used to describe the mechanical and thermal properties of the constructed mechanical components. The multi-physics simulation model of the i-th mechanical component can be expressed as

[0081] M model(i) =M(G model(i) , M PM(i) , L M(i) )

[0082] T model(i) =T(G model(i) , M PT(i) , L T(i) )

[0083] Among them, M model(i)represents the mechanical model of the i-th component, T model(i) represents the thermal model of the ith component, M PM(i) represents the mechanical material properties of the i-th component, M PT(i) represents the thermal material properties of the ith component, L M(i) represents the force load of the ith component, L T(i) represents the thermal load of the i-th component.

[0084] Construct a simulation model of mechanical components. The simulation model includes a fatigue life model, a failure rate model, and a reliability model, which are used to describe the fatigue life characteristics, failure rate characteristics, and reliability characteristics of the constructed mechanical components. The fatigue life model of the i-th mechanical component can be expressed as

[0085] L3 model(i) (L (i) )=L(M model(i) , T model(i) , C F(i) , L S(i) )

[0086] Among them, L3 model(i) (L (i) ) represents the simulation-based fatigue life model of the i-th component, C F(i) represents the surface coefficient set related to the manufacturing process of the i-th component, L S(i) represents the full life cycle load spectrum of the i-th component.

[0087] In order to obtain the failure rate model and reliability model of mechanical components, it is necessary to discretize the input parameters of the fatigue life model to obtain a series of corresponding output fatigue life values ​​to form a sample space. The fatigue life of mechanical components often obeys the Weibull distribution, that is, the failure probability density function based on the three-parameter Weibull distribution is: Therefore, the failure rate model and reliability model of mechanical components are obtained as follows:

[0088]

[0089] Among them, λ3 model(i) represents the simulation-based failure rate model of the ith component, R3 model(i) represents the simulation-based reliability model of the i-th component, t represents the operating time of the mechanical component, m, η and γ represent the shape parameter, scale parameter and location parameter, respectively, which are obtained by fitting the Weibull distribution.

[0090] Based on the above simulation model construction, the simulation model of the i-th mechanical component can be expressed as

[0091] S model(i) (L3 (i) ,λ3(i) , R3 (i) )=S(L3 model(i) ,λ3 model(i) , R3 model(i) )

[0092] Among them, S model(i) (L3 (i) ,λ3 (i) , R3 (i) ) represents the simulation model of the i-th mechanical component, L3 (i) represents the life parameter of the output of the i-th mechanical component, λ3 (i) represents the failure rate parameter of the output of the i-th mechanical component, R3 (i) Represents the reliability parameter of the output of the i-th mechanical component.

[0093] After building the simulation model of the mechanical component, it is necessary to build a reliability digital twin model of the mechanical component. The construction method is similar to the digital twin model of the mechanical component based on the failure mechanism model. The steps are as follows:

[0094] Using the Simulink module in Matlab software, the obtained simulation model mathematical expression of the i-th mechanical component is used to create a simulation model in the Simulink environment, the step mode and step value of the model solution are configured, and the i-th mechanical component functional model unit FMU3 that meets the FMI standard is exported. (i) ,Right now

[0095]

[0096] Among them, FMU3 (i) represents the functional model unit based on the simulation model of the i-th mechanical component, The simulation model of the i-th mechanical component is formed through simulation and conforms to the FMI protocol.

[0097] Since this FMU3 (i) It has fast computing capability and a universal FMI interface for system simulation, so it can be used as a reliability digital twin model based on the simulation model for the i-th mechanical component.

[0098] RDT3 (i) =FMU3 (i)

[0099] Among them, RDT3 (i) Represents the reliability digital twin model of the i-th mechanical component based on the simulation model.

[0100] The fourth category: Construction of digital twin models of mechanical components based on structural functional functions

[0101] The fourth type of mechanical component digital twin model based on structure-function function is built on the premise of building its fatigue simulation model first. The structure-function function model is established through joint simulation and limit state equation. The steps are as follows:

[0102] To construct the functional structure function, we must first do simulation, and then build the structure function function based on the simulation. Therefore, similar to the construction of the previous simulation model, we will build the geometric model G of the mechanical parts in turn. model(i) 、Mechanical model M model(i) and thermal model T model(i) .

[0103] Then, a structural function simulation model of the mechanical component is constructed. The structural function simulation model includes a fatigue life model, a failure rate model, and a reliability model, which are used to describe the fatigue life characteristics, failure rate characteristics, and reliability model of the constructed mechanical component. The fatigue life model of the i-th mechanical component can be expressed as

[0104] L4 model(i) (L (i) )=L(M model(i) , Y model(i) , C F(i) , L S(i) )

[0105] Among them, L4 model(i) represents the fatigue life model based on simulation, C F(i) represents the surface coefficient set related to the manufacturing process of the i-th component, L S(i) represents the full life cycle load spectrum of the i-th component.

[0106] Construct the structure-function function of mechanical components. The mechanical, thermal and life structure-function functions of the i-th mechanical component are expressed as

[0107] SFF Mmodel(i) (M (i) )=M model(i) -M AM(i)

[0108] SFF Tmodel(i) (T (i) )=T model(i) -T AT(i)

[0109] SFF Lmodel(i) (L (i) )=L4 model(i) -L AL(i)

[0110] Among them, SFF Mmodel(i) (M (i)) represents the mechanical structure function of the i-th mechanical component, SFF Tmodel(i) (T (i) ) represents the thermal-based structure function of the i-th mechanical component, SFF Lmodel(i) (L (i) ) represents the structural function based on fatigue life of the i-th mechanical component, M AM(i) represents the mechanical structural resistance of the i-th component, T AT(i) represents the thermal structural resistance of the ith component, L AL(i) represents the life structural resistance of the ith component.

[0111] In order to obtain the failure rate model and reliability model of mechanical components, the input parameters of the function function need to be discretized, so that a series of corresponding output values ​​of the output mechanical, thermal and life structure function functions can be obtained to form the response sample space. The mean value first order second moment method (MVFOSM) is used to calculate the failure rate and reliability. The calculation process is as follows:

[0112] Suppose the structural function SSF = g(X1, X2, ..., X n ), where X1, X2, …, X n is a basic random variable, representing various uncertain factors that affect structural performance, such as material strength, load size, etc. First, determine the mean value μ of the basic random variable x , that is, call (X i ). Then, calculate the performance function SSF at the mean point The partial derivatives of each random variable According to Taylor series expansion and ignoring high-order terms, the approximate expression of the performance function at the mean point is: Since the mean of the random variable Known, and when , the subsequent summation term is zero, so the mean of the function is The variance is

[0113] in is a random variable X i The standard deviation of ij is a random variable X i With X j The correlation coefficient between them. Then, the failure rate model and reliability model of the structure are:

[0114]

[0115] Among them, λ4 model(i)represents the failure rate model based on the structural performance function of the i-th component, R4 model(i) represents the reliability model of the i-th component based on the structural performance function, μ SSF and σ SSF They represent the mean and variance of the normal distribution sample points, respectively, which are obtained by normal distribution fitting, and Φ[·] represents the standard normal distribution function.

[0116] Based on the above structure-function function model construction, the structure-function function model of the i-th mechanical component can be expressed as

[0117] SFF model(i) (L4 (i) ,λ4 (i) , R4 (i) )=SFF(L4 model(i) ,λ4 model(i) , R4 model(i) )

[0118] Among them, SFF model(i) (L4 (i) ,λ4 (i) , R4 (i) ) represents the structural function model of the i-th mechanical component, L4 (i) represents the life parameter of the output of the i-th mechanical component, λ4 (i) represents the failure rate parameter of the output of the i-th mechanical component, R4 (i) Represents the reliability parameter of the output of the i-th mechanical component.

[0119] After building the structural function model of the mechanical component, it is necessary to build a reliability digital twin model of the mechanical component. The construction method is similar to the digital twin model of the mechanical component based on the failure mechanism model. The steps are as follows:

[0120] Using the Simulink module in Matlab software, the mathematical expression of the structure function model of the i-th mechanical component is created in the Simulink environment for simulation model creation, the step mode and step value of the model solution are configured, and the functional model unit FMU4 of the i-th mechanical component that meets the FMI standard is exported. (i) ,Right now

[0121]

[0122] Among them, FMU4 (i) represents the functional model unit based on the structure-function function model of the i-th mechanical component, The structure function function model of the i-th mechanical component is formed through simulation and conforms to the FMI protocol.

[0123] Since this FMU4(i) It has fast computing capability and a universal FMI interface for system simulation, so it can be used as a reliability digital twin model of the i-th mechanical component based on the structure-function function model.

[0124] RDT4 (i) =FMU4 (i)

[0125] Among them, RDT4 (i) Represents the reliability digital twin model of the i-th mechanical component based on the structure-functional function model.

[0126] Fifth category: Construction of digital twin models of mechanical components based on data models

[0127] The premise for building the fifth type of digital twin model of mechanical components based on data models is to obtain a large amount of data that can reflect the mapping relationship between the input parameters of mechanical components and reliability indicators such as life, failure rate, and reliability. This data model is mainly built through historical data or data obtained through experimental tests. After obtaining the reliability data, the data model is built through the proxy model method. The steps are as follows:

[0128] First, we construct the input sample space based on the data. Assuming that the sample space of the data contains n independent variables, each independent variable has m sets of data, and the output sample space contains three response values, namely life, failure rate, and reliability, the sample space of the data can be expressed as:

[0129]

[0130] in, Represents an m×n dimensional array space consisting of n independent variables, Represents the m×3 dimensional array space composed of response value life, failure rate, and reliability.

[0131] Based on the Kriging reduction method, the input and output sample space is trained by minimizing the error or maximizing the likelihood estimation, and the following relationship between the response value and the independent variable is obtained:

[0132] y(X)=f T (X)β+z(X)

[0133] Where, f(X)=[f1(X),f2(X),…,f n (X)] T is a known regression model, usually a polynomial function; β is the corresponding unknown parameter; f T (X)β is a deterministic part, called deterministic drift; z(X) is called fluctuation, which is a statistical process with a mean of 0 and a variance of

[0134] At this point, the reduced-order models of life, failure rate, and reliability can be obtained, which are

[0135]

[0136] in, represents the life reduction model of the i-th component, represents the reduced-order model of the failure rate of the i-th component, represents the reliability reduction model of the i-th component, (x1,…x i , …x n ) represents the input sample parameters.

[0137] Based on the above data function model construction, the data model of the i-th mechanical component can be expressed as

[0138]

[0139] Among them, D model(i) (L5 (i) ,λ5 (i) , R5 (i) ) represents the data model of the i-th mechanical component, L5 (i) represents the life parameter of the output of the i-th mechanical component, λ5 (i) represents the failure rate parameter of the output of the i-th mechanical component, R5 (i) Represents the reliability parameter of the output of the i-th mechanical component.

[0140] After building the data model of the mechanical components, it is necessary to build a reliability digital twin model of the mechanical components. The construction method is similar to the digital twin model of the mechanical components based on the failure mechanism model. The steps are as follows:

[0141] Using the Simulink module in Matlab software, the mathematical expression of the data model of the i-th mechanical component is used to create a simulation model in the Simulink environment, the step mode and step value of the model solution are configured, and the i-th mechanical component functional model unit FMU5 that meets the FMI standard is exported. (i) ,Right now

[0142]

[0143] Among them, FMU5 (i) represents the functional model unit based on the data model of the i-th mechanical component, The data model representing the i-th mechanical component is a data model that complies with the FMI protocol after simulation.

[0144] Since this FMU5 (i)It has fast computing capability and a universal FMI interface for system simulation, so it can be used as a reliability digital twin model based on a data model for the i-th mechanical component.

[0145] RDT5 (i) =FMU5 (i)

[0146] Among them, RDT5 (i) Represents the reliability digital twin model of the i-th mechanical component based on the data model.

[0147] To sum up, the method of the embodiment of the present invention focuses on reliability. By establishing a digital twin model of the reliability of mechanical components, it realizes the simulation, monitoring, diagnosis, prediction and control of reliability indicators such as failure rate, reliability, life, and remaining life of mechanical components, thereby using digital means to realize reliability design analysis, test evaluation and diagnostic prediction of mechanical components, greatly shortening the product iteration and upgrade cycle, and greatly reducing the research and development and use costs of mechanical components.

[0148] An embodiment of the present invention further provides a computer device, which may be a server, and its internal structure diagram may be as shown in FIG. Figure 3 As shown. The computer device includes a processor, a memory and a network interface connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store operating parameter data of each framework. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the steps of the method for constructing a digital twin model of mechanical component reliability according to an embodiment of the present invention are implemented.

[0149] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0150] An embodiment of the present invention also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method for constructing a digital twin model of mechanical component reliability according to an embodiment of the present invention are implemented.

[0151] An embodiment of the present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the method for constructing a digital twin model of mechanical component reliability according to an embodiment of the present invention.

[0152] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for constructing a digital twin model of mechanical component reliability, characterized in that: include: Step S1: for the i-th mechanical component, determine whether there is a failure mechanism model. If the determination is no, execute step S2. If the determination is yes, construct a digital twin model of the mechanical component based on the failure mechanism model and execute step S7: Among them, RDT1 (i) represents the reliability digital twin model of the i-th mechanical component based on the failure mechanism model, The failure mechanism model of the i-th mechanical component is formed by simulation and conforms to the FMI protocol. L1 (i) represents the life parameter of the output of the i-th mechanical component, λ1 (i) Represents the failure rate parameter of the i-th mechanical component output, R1 (i) represents the reliability parameter of the output of the i-th mechanical component; Step S2, determine whether there is an empirical model. If the judgment is no, execute step S3. If the judgment is yes, build a digital twin model of the mechanical component based on the empirical model and execute step S7: Among them, RDT2 (i) represents the reliability digital twin model of the i-th mechanical component based on the empirical model, The empirical model of the i-th mechanical component is formed by simulation and conforms to the FMI protocol. L2 (i) represents the life parameter of the output of the i-th mechanical component, λ2 (i) represents the failure rate parameter of the output of the i-th mechanical component, R2 (i) represents the reliability parameter of the output of the i-th mechanical component; Step S3, determine whether simulation is possible, if it is judged as no, execute step S4, if it is judged as yes, build a digital twin model of the mechanical component based on the simulation model and execute step S7: Among them, RDT3 (i) represents the reliability digital twin model of the i-th mechanical component based on the simulation model, The simulation model of the i-th mechanical component is a simulation model that complies with the FMI protocol after simulation. L3 (i) represents the life parameter of the output of the i-th mechanical component, λ3 (i) represents the failure rate parameter of the output of the i-th mechanical component, R3 (i) represents the reliability parameter of the output of the i-th mechanical component; Step S4, determine whether there is a structure-function function. If it is determined to be no, execute step S5. If it is determined to be yes, construct a digital twin model of the mechanical component based on the structure-function function model and execute step S7: Among them, RDT4 (i) represents the reliability digital twin model of the i-th mechanical component based on the structure-function function model, The structure function function model of the i-th mechanical component is formed by simulation and conforms to the FMI protocol. L4 (i) represents the life parameter of the output of the i-th mechanical component, λ4 (i) represents the failure rate parameter of the output of the i-th mechanical component, R4 (i) represents the reliability parameter of the output of the i-th mechanical component; Step S5, determine whether there is historical data. If yes, build a digital twin model of the mechanical component based on the data model and execute step S7: Among them, RDT5 (i) represents the reliability digital twin model of the i-th mechanical component based on the data model, The data model of the i-th mechanical component is simulated to form a data model that conforms to the FMI protocol, L5 (i) represents the life parameter of the output of the i-th mechanical component, λ5 (i) represents the failure rate parameter of the output of the i-th mechanical component, R5 (i) represents the reliability parameter of the output of the i-th mechanical component; Step S6, if the judgment in step S5 is no, obtain the test data as the historical data through the test and repeat step S5; Step S7, judging whether the modeling of all mechanical components is completed, if it is judged as no, returning to step S1 to model the next mechanical component until the reliability modeling of all mechanical components is completed.

2. The method according to claim 1, characterized in that In step S1, the failure mechanism model of the i-th mechanical component is expressed as: FM model(i) (L1 (i) ,λ1 (i) ,R1 (i) )=M1(G P(i) ,O P(i) ,P P(i) ,S P(i) ) Among them, FM model(i) (L1 (i) ,λ1 (i) ,R1 (i) ) represents the failure mechanism model of the i-th mechanical component, G P(i) represents the geometric model parameters of the i-th mechanical component, O P(i) represents the operating state parameter of the i-th mechanical component, P P(i) represents the external load parameter of the i-th mechanical component, S P(i) Represents the structural characteristic parameters of the i-th mechanical component.

3. The method according to claim 2, characterized in that The geometric model parameters of the i-th mechanical component are expressed as: G P(i) G1(D (i) ,AM (i) ) Among them, D (i) represents the size parameter of the i-th mechanical component, A (i) represents the assembly parameters of the i-th mechanical component; The operating state parameter of the i-th mechanical component is expressed as: O P(i) =O(R P(i) ,T P(i) ,F P(i) ,C P(i) ,V P(i) ) Among them, R P(i) represents the speed parameter of the i-th mechanical component, T P(i) represents the torque parameter of the i-th mechanical component, F P(i) represents the flow parameter of the i-th mechanical component, C P(i) represents the current parameter of the i-th mechanical component, V P(i) represents the voltage parameter of the i-th mechanical component; The external load parameters of the i-th mechanical component are expressed as: P P(i) =P(L M(i) ,L T(i) ,L H(i) ) Among them, L M(i) represents the force load parameter of the i-th mechanical component, L T(i) represents the thermal load parameter of the i-th mechanical component, L H(i) represents the humidity load parameter of the i-th mechanical component; The structural characteristic parameters of the i-th mechanical component are expressed as: S P(i) =S(M C(i) ,P C(i) ,K C(i) ,E C(i) ,C C(i) ) Among them, M C(i) represents the material coefficient of the i-th mechanical component, P C(i) Represents the process coefficient of the i-th mechanical component, K C(i) represents the motion coefficient of the ith mechanical component, E C(i) represents the electrical stress coefficient of the i-th mechanical component, C C(i) represents the constant coefficient of the i-th mechanical component.

4. The method according to claim 2 or 3, characterized in that: In step S2, the empirical model of the i-th mechanical component is expressed as: E model(i) (L2 (i) ,λ2 (i) ,R2 (i) )=E(G P(i) ,O P(i) ,P P(i) ,S P(i) ) Among them, E model(i) (L2 (i) ,λ2 (i) ,R2 (i) ) represents the empirical model of the i-th mechanical component.

5. The method according to claim 1 or 2, characterized in that: In step S3, the simulation model of the i-th mechanical component is represented by: S model(i) (L3 (i) ,λ3 (i) ,R3 (i) )=S(L3 model(i) ,λ3 model(i) ,R3 model(i) ) Among them, S model(i) (L3 (i) ,λ3 (i) ,R3 (i) ) represents the simulation model of the i-th mechanical component, L3 model(i) =L(M model(i) ,T model(i) ,C F(i) ,L S(i) ) Among them, L3 model(i) represents the simulation-based fatigue life model of the ith component, C F(i) represents the surface coefficient set related to the manufacturing process of the i-th component, L S(i) represents the full life cycle load spectrum of the i-th component, M model(i) =M(G model(i) ,M PM(i) ,L M(i) ) T model(i) =T(G model(i) ,M PT(i) ,L T(i) ) Among them, M model(i) represents the mechanical model of the i-th component, T model(i) represents the thermal model of the ith component, M PM(i) represents the mechanical material properties of the i-th component, M PT(i) represents the thermal material properties of the ith component, L M(i) represents the force load of the ith component, L T(i) represents the thermal load of the ith component, G model(i) =G2(D (i) ,A (i) ) Among them, G model(i) represents the geometric model of the i-th mechanical component, D (i) represents the size parameter of the i-th mechanical component, A (i) represents the assembly parameters of the i-th mechanical component, Among them, λ3 model(i) represents the simulation-based failure rate model of the ith component, R3 model(i) represents the simulation-based reliability model of the i-th component, t represents the operating time of the mechanical component, m, η, and γ represent the shape parameter, scale parameter, and location parameter, respectively, which are obtained by fitting the Weibull distribution.

6. The method according to claim 5, characterized in that In step S4, the structure-function function model of the i-th mechanical component is expressed as: SFF model(i) (L4 (i) ,λ4 (i) ,R4 (i) )=SFF(L4 model(i) ,λ4 model(i) ,R4 model(i) ) Among them, SFF model(i) (L4 (i) ,λ4 (i) ,R4 (i) ) represents the structure-function function model of the i-th mechanical component; L4 model(i) =L(M model(i) ,T model(i) ,C F(i) ,L S(i) ) Among them, L4 model(i) represents the simulation-based fatigue life model, Among them, λ4 model(i) represents the failure rate model based on the structural performance function of the i-th component, R4 model(i) represents the reliability model of the i-th component based on the structural performance function, μ SSF and σ SSF They represent the mean and variance of the normal distribution sample points, respectively, which are obtained by normal distribution fitting, and Φ[·] represents the standard normal distribution function.

7. The method according to claim 1 or 2, characterized in that: In step S5, the data model of the i-th mechanical component is expressed as: Among them, D model(i) (L5 (i) ,λ5 (i) ,R5 (i) ) represents the data model of the i-th mechanical component, represents the life reduction model of the i-th component, represents the reduced-order model of the failure rate of the i-th component, represents the reliability reduced-order model of the i-th component.

8. The method according to claim 7, characterized in that The input sample space is constructed based on the data. Assume that the input sample space of the data contains n independent variables, each independent variable has m groups of data, and the output sample space contains three response values, namely life, failure rate, and reliability. Based on the Kriging reduction method, the input and output sample space is trained by minimizing the error or maximizing the likelihood estimation, and the relationship between the response value and the independent variable is obtained: y(X)=f T (X)β+z(X) Where X represents the m×n dimensional array space composed of n independent variables, y represents the response value, f(X)=[f1(X),f2(X),…,f n (X)] T is a regression model, β is an undetermined parameter, f T (X)β is the deterministic drift, z(X) is the fluctuation, then: Among them, x1,…x i , …x n Represents the input sample parameters.

9. A computer device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images