Method and system for evaluating service life of SIS logic controller

Abstract

The invention discloses a method and system for evaluating the service life of an SIS logic controller, and the method comprises the steps: generating an initial sample space according to all factors affecting the working condition of the SIS logic controller; loading the samples under each combination factor on a preset finite element model to obtain a first life result, and then comparing the first life result with a result of an accelerated life experiment under the same condition to obtain a life error; establishing a data driving model representing the relationship between the combined factor input parameter and the life error parameter, and solving the data driving model to obtain an optimal combined factor data set; obtaining second life data of the optimal combination factor data set under different input current conditions, and constructing a life evaluation model by using different input current data and the corresponding second life data; and inputting current data, and obtaining a dynamic life evaluation result by using the life evaluation model. According to the method, the service life model suitable for the SIS logic controller is established, and the integrity evaluation capability of the SIS logic controller is formed.

Description

Technical Field

[0001] The present invention relates to the field of high-precision evaluation of reliability of SIS logic controllers, and in particular to a method and system for evaluating the life of SIS logic controllers. Background Art

[0002] As my country's industrial process continues to expand, the scale of production in the chemical industry has also increased. Against this background, the field of chemical safety is facing more severe challenges. Safety instrument system (SIS) has the advantages of wide coverage, high safety, and self-diagnosis. In the SIS controller, it is mainly composed of hardware parts such as core PU board, DI, DO, AI, and AO board. The failure of any board will affect the functional integrity of the entire SIS. However, since SIS has entered the chemical industry for a short time, research on its reliability is currently mainly concentrated in the design and development stage, and is mainly focused on manufacturers. Considering that petrochemical and refining scenes are usually in harsh outdoor environments, the operating reliability and service life of SIS controllers are greatly reduced. Therefore, it is of positive significance to carry out hardware reliability model analysis and evaluation of safety instrument SIS systems in petrochemical scenarios.

[0003] Since SIS logic controllers have only been introduced to China for a short time, targeted research on the reliability of this object has not been fully carried out. In related industries, such as conventional DCS and PLC controllers, the main research methods include finite element numerical simulation and accelerated aging experiments based on state physics equations. Studies have shown that the main cause of failure is the failure of board integrated circuits, so existing technologies are mostly focused on the thermal reliability and thermal stress reliability of integrated circuit boards. In finite element technology, in order to reduce the amount of calculation, research often uses boundary settings under ideal conditions, and in order to avoid the non-convergence problems caused by high-dimensional and complex physical state equations, factors such as salt spray, electromagnetic interference, and environmental humidity are often ignored. However, in the harsh petrochemical background, the above factors will lead to a reduction in the life of the SIS system, which is far from the actual experimental data and the life results under real working conditions. This problem not only plagues the SIS system, but is also a thorny problem facing the current electronics industry.

[0004] On the other hand, some reliability assessments of electronic products often use classic aging experiments. In the environmental laboratory, loading conditions similar to the working conditions are provided to carry out accelerated aging experiments on the controller. Subsequently, the experimental data is trained through intelligent algorithms such as Monte Carlo and neural networks to fit the correlation coefficients of the accelerated life model, and then the actual life under normal working conditions is inferred. However, the precise fitting coefficient is closely related to the experimental samples. If you want to obtain an accurate model using this method, a large number of experimental samples are required, which consumes a lot of time and financial resources, greatly increasing the economic cost.

[0005] In view of the problems existing in traditional technologies, it is urgent to organically combine finite element numerical simulation and aging experiments, use computer software to optimize the errors between numerical simulation and experimental results, obtain more and more accurate experimental samples, and establish a life evaluation model suitable for SIS logic controllers, so as to form a more efficient and accurate reliability analysis method for SIS logic controllers, thereby scientifically and completely evaluating the life results of SIS logic controllers. Summary of the invention

[0006] The purpose of the present invention is to address the problem that the actual experimental data is too different from the life results under real working conditions due to the traditional technology ignoring various working conditions factors, and to propose a technical solution for evaluating the life of SIS logic controller by establishing a life evaluation model.

[0007] In order to solve the above technical problems, an embodiment of the present invention provides a method for evaluating the life of a SIS logic controller, comprising: step 1, generating an initial sample space according to various factors affecting the working conditions of the SIS logic controller; step 2, loading samples under various combination factors in the initial sample space on a preset finite element model to obtain a first life result under the corresponding combination factor, and then comparing the first life result with the result of an accelerated life test under the same conditions to obtain a life error; step 3, establishing a data-driven model that characterizes the relationship between the input parameters of the combination factors and the life error parameters, and solving the data-driven model according to the samples of each combination factor and the corresponding life error to obtain an optimal combination factor data group; step 4, obtaining the second life data of the optimal combination factor data group under different input current conditions, thereby constructing a life evaluation model suitable for the SIS logic controller based on the basic failure physics model and using different input current data and the corresponding second life data; step 5, using the life evaluation model according to the real-time input current data of the SIS logic controller to be evaluated, to obtain a dynamic life evaluation result.

[0008] Preferably, in the method, step one includes: determining various variable factors and corresponding parameter ranges; filling the sample space with the optimal Latin hypercube criterion according to the parameter ranges of various variable factors to form a three-dimensional sample space distribution; and normalizing the three-dimensional sample space distribution to obtain the initial sample space.

[0009] Preferably, the method, in the process of generating the first life result, includes: constructing a three-dimensional finite element model of the SIS logic controller, and loading the rated current on the current controller, and then sequentially loading the samples under each combination factor in the initial sample space onto the three-dimensional finite element model, combining with the linear cumulative damage theory, and taking the controller chip temperature exceeding the preset threshold temperature as the cutoff basis, to obtain the first life result under different sample loading conditions.

[0010] Preferably, the method, in the process of solving the data-driven model, includes: continuously adjusting the factor data in the initial sample space so that the life error parameter output by the model reaches a minimum value, thereby obtaining the optimal combination factor data group.

[0011] Preferably, the method, in step three, includes: a first step, fitting the data-driven model according to each combination factor sample and the corresponding life error to obtain an initial relationship model; a second step, obtaining a new input sample point based on a mixed point addition criterion, and using the initial relationship model to obtain a response value of the current new sample point; a third step, judging whether the current relationship model meets the application requirements according to the new sample point and the corresponding response value and using a preset convergence judgment condition; if so, taking the numerical range of each factor corresponding to the minimum life value of the current relationship model as the optimal combination factor data group.

[0012] Preferably, if the current relationship model does not meet the application requirements, the method adds new sample points based on the point adding criteria and obtains corresponding response values ​​using the current initial relationship model, thereby returning to the first step to update the initial relationship model based on the samples of each combination factor and the corresponding life error, as well as the newly added sample points and their response values.

[0013] Preferably, the method and the data-driven model are implemented using a Kriging proxy model.

[0014] Preferably, the method, in step four, includes: based on the three-dimensional finite element model of the SIS logic controller, and using the optimal combination factor data group as the controller operating environment condition; applying multiple groups of input current data to the controller in the finite element model to obtain second life data under different input current sample conditions; based on the basic failure physics model, according to different input current samples and the corresponding second life data, obtaining the life evaluation model that characterizes the relationship between current and life.

[0015] Preferably, the method, in the step of obtaining the life evaluation model characterizing the relationship between current and life based on the basic failure physics model according to different input current samples and the corresponding second life data, includes: obtaining a training data group and a verification data group according to the input current samples and the corresponding second life data samples in accordance with a preset ratio of the training data amount to the verification data amount, wherein the ratio of the training data amount to the verification data amount is 4:1; fitting the basic failure physics model according to the training data group to obtain an initial life prediction model containing initial fitting coefficients; and verifying the initial life prediction model according to the verification data group to update the initial fitting coefficients to obtain the life evaluation model containing the updated fitting coefficients.

[0016] Preferably, in the method, the basic failure physics model is expressed using the following expression:

[0017]

[0018] Among them, t represents the life parameter, E represents the failure acceleration energy, k represents the Boltzmann constant, i represents the current parameter, A, They represent the coefficients to be fitted.

[0019] On the other hand, the present invention also proposes a computer-readable storage medium, which contains a series of instructions for executing the method steps described above.

[0020] On the other hand, the present invention also proposes a system for evaluating the life of a SIS logic controller, comprising: an influencing factor sample space generation module, which is configured to generate an initial sample space according to various factors affecting the working conditions of the SIS logic controller; a life error sample data generation module, which is configured to load samples under various combination factors in the initial sample space on a preset finite element model to obtain a first life result under the corresponding combination factor, and then compare the first life result with the result of an accelerated life experiment under the same conditions to obtain a life error; an optimal combination factor generation module, which is configured to establish a data-driven model that characterizes the relationship between the combination factor input parameter and the life error parameter, and solve the data-driven model according to each combination factor sample and the corresponding life error to obtain an optimal combination factor data group; a life evaluation model generation module, which is configured to obtain the second life data of the optimal combination factor data group under different input current conditions, so as to construct a life evaluation model suitable for the SIS logic controller based on the basic failure physics model and using different input current data and the corresponding second life data; an online evaluation module, which is configured to obtain a dynamic life evaluation result according to the real-time input current data of the SIS logic controller to be evaluated using the life evaluation model.

[0021] Compared with the prior art, one or more embodiments of the above scheme may have the following advantages or beneficial effects:

[0022] The present invention proposes a method and system for evaluating the life of SIS logic controllers. The present invention solves the problem that the traditional technology ignores multiple environmental factors and causes the life result data to be inconsistent with the actual situation. The present invention considers the main environmental variable characteristics that cause the failure of the SIS logic controller, and compares the life results of finite element modeling and accelerated experiments; and on the basis of the traditional integrated circuit thermal stress finite element model, multiple boundary conditions are loaded, and the boundary condition parameters are optimized using the proxy model to make the environmental variables closer to the actual working conditions, thereby establishing a life evaluation model suitable for the SIS logic controller, forming a complete life evaluation capability for the SIS logic controller.

[0023] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0025] Figure 1 It is a schematic diagram of the steps of the method for evaluating the life of a SIS logic controller according to an embodiment of the present application.

[0026] Figure 2 The present invention is a flowchart of the step of determining an initial sample space in the method for evaluating the life of a SIS logic controller according to an embodiment of the present application.

[0027] Figure 3 This is an example diagram of the spatial distribution of initial sample points in the method for evaluating the life of a SIS logic controller according to an embodiment of the present application.

[0028] Figure 4 The present invention is a flow chart of the step of optimizing the boundary conditions of the finite element model in the method for evaluating the life of the SIS logic controller according to an embodiment of the present application.

[0029] Figure 5 This is a flow chart of constructing a SIS logic controller failure physical equation in a method for evaluating the life of a SIS logic controller according to an embodiment of the present application.

[0030] Figure 6 It is a schematic diagram of the overall structure of the system for evaluating the life of a SIS logic controller according to an embodiment of the present application. DETAILED DESCRIPTION

[0031] The following will describe the implementation methods of the present invention in detail with reference to the accompanying drawings and embodiments, so that the implementation process of how the present invention applies technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly. It should be noted that as long as there is no conflict, the various embodiments of the present invention and the various features in the embodiments can be combined with each other, and the technical solutions formed are all within the protection scope of the present invention.

[0032] In addition, the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions. Also, although a logical sequence is shown in the flowchart, in some cases, the steps shown or described can be performed in a sequence different from that here.

[0033] The terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments. Unless the context clearly indicates otherwise, the singular forms "one", "one" and "item" used herein are also intended to include plural numbers. It should also be understood that the terms "include" and / or "comprise" used herein specify the existence of stated features, integers, steps, operations, units and / or components, without excluding the existence or addition of one or more other features, integers, steps, operations, units, components and / or combinations thereof.

[0034] As my country's industrial process continues to expand, the scale of production in the chemical industry has also increased. Against this background, the field of chemical safety is facing more severe challenges. Safety instrument system (SIS) has the advantages of wide coverage, high safety, and self-diagnosis function. Since 2018, the state has proposed that all chemical plants involving "two key and one major" must design corresponding SIS devices. In the SIS controller, it is mainly composed of core PU board, DI, DO, AI, AO board and other hardware parts. The failure of any board will affect the functional integrity of the entire SIS. However, since SIS has entered the chemical industry for a short time, the research on its reliability is currently mainly concentrated in the design and development stage, and mainly by manufacturers. Considering that petrochemical and refining scenes are usually in harsh outdoor environments, the operating reliability and service life of SIS controllers are greatly reduced. Therefore, it is of positive significance to carry out hardware reliability model analysis and evaluation of safety instrument SIS systems in petrochemical scenes.

[0035] Since SIS logic controllers have only been introduced to China for a short time, targeted research on the reliability of this object has not been fully carried out. In related industries, such as conventional DCS and PLC controllers, the main research methods include finite element numerical simulation and accelerated aging experiments based on state physics equations. Studies have shown that the main cause of failure is the failure of board integrated circuits, so existing technologies are mostly focused on the thermal reliability and thermal stress reliability of integrated circuit boards. In finite element technology, in order to reduce the amount of calculation, research often uses boundary settings under ideal conditions, and in order to avoid the non-convergence problems caused by high-dimensional and complex physical state equations, factors such as salt spray, electromagnetic interference, and environmental humidity are often ignored. However, in the harsh petrochemical background, the above factors will lead to a reduction in the life of the SIS system, which is far from the actual experimental data and the life results under real working conditions. This problem not only plagues the SIS system, but is also a thorny problem facing the current electronics industry.

[0036] On the other hand, some reliability assessments of electronic products often use classic aging experiments. In the environmental laboratory, loading conditions similar to the working conditions are provided to carry out accelerated aging experiments on the controller. Subsequently, the experimental data is trained through intelligent algorithms such as Monte Carlo and neural networks to fit the correlation coefficients of the accelerated life model, and then the actual life under normal working conditions is inferred. However, the precise fitting coefficient is closely related to the experimental samples. If you want to obtain an accurate model using this method, a large number of experimental samples are required, which consumes a lot of time and financial resources, greatly increasing the economic cost.

[0037] In view of the problems existing in traditional technologies, it is urgent to organically combine finite element numerical simulation and aging experiments, use computer software to optimize the errors between numerical simulation and experimental results, obtain more and more accurate experimental samples, and establish a life evaluation model suitable for SIS logic controllers, so as to form a more efficient and accurate reliability analysis method for SIS logic controllers, thereby scientifically and completely evaluating the life results of SIS logic controllers.

[0038] In order to solve the above problems, the present invention proposes a solution for evaluating the life of a SIS logic controller. The solution takes into account the environmental factors that affect the SIS logic controller, integrates the environmental factors into the physical equations, and generates a sample space based on the variables of the environmental factors to carry out the modeling process of the finite element model. In addition, multiple boundary conditions are loaded on the basis of the traditional integrated circuit thermal stress finite element model, and the boundary condition parameters are optimized using the proxy model to make the environment in which the model is located closer to the actual working conditions, so that the life model of the SIS logic controller established can better guide the actual situation. And using the life evaluation model, a dynamic life evaluation result is generated to form a complete SIS logic controller life evaluation capability.

[0039] Embodiment 1

[0040] Figure 1 This is a schematic diagram of the steps of the method for evaluating the life of a SIS logic controller according to an embodiment of the present application. Figure 1 The method for evaluating the life of a SIS logic controller in the present invention is described in detail.

[0041] like Figure 1 As shown, step S110 generates an initial sample space according to various factors affecting the operating conditions of the SIS logic controller.

[0042] In one embodiment, Figure 2 As shown, in step S110 of generating the sample space, it includes: first, determining various variable factors affecting the operating conditions of the SIS logic controller and the parameter range corresponding to each variable factor; then, according to the parameter range of each variable factor, using the optimal Latin hypercube criterion to fill the sample space to form a three-dimensional sample space distribution; finally, normalizing the three-dimensional sample space distribution to obtain the initial sample space.

[0043] Specifically, firstly, according to the field working conditions of the petrochemical industry, the characteristics of the external environmental condition variables affecting the SIS logic controller and the numerical range of each characteristic parameter are fully considered. Then, the Latin hypercube sampling method is used to sample the environmental variables within the environmental parameter range to form a three-dimensional sample space distribution. Finally, the three-dimensional sample space is optimized to obtain the initial sample space X, see Figure 3 In the embodiment of the present invention, the main environmental variables affecting the SIS logic controller are factors such as high temperature in summer, humidity in the air, and vibration caused by large moving equipment.

[0044] After the initial sample space is obtained, the process proceeds to step S120. In step S120, the samples under different combination factors in the initial sample space obtained in step S110 are loaded on a preset finite element model to obtain the first life result under the corresponding combination factors, and then the first life result is compared with the result of the accelerated life test under the same conditions to obtain the life error.

[0045] In one embodiment, Figure 4 As shown, the steps of generating the first life result include: constructing a three-dimensional finite element model of the SIS logic controller, and loading the rated current on the current controller, and then sequentially loading the samples under various combination factors in the initial sample space constructed according to the Latin hypercube on the three-dimensional finite element model, combining the linear cumulative damage theory, and taking the controller chip temperature exceeding the preset threshold temperature as the cutoff basis, to obtain the first life result under different sample loading conditions.

[0046] According to the characteristics of electronic components, their failure is mostly due to the accumulation of Joule heat under the electro-thermal effect after the chip has been powered on for a long time, resulting in chip breakdown. Therefore, when loading factors, the variable that needs to be input is the current, so the rated current is applied to the board chip. Then, each combination factor of the initial sample space is loaded on the finite element model in turn. Because the damage to the finite element model caused by each sample in each combination factor is cumulative, loading each group of factors on the finite element model will obtain different life results. Therefore, when loading different combination factors, the Comsol fatigue module will be used to carry out multi-physics field simulation to obtain the SIS logic controller finite element simulation life result (i.e., the first life result), so as to compare the first life result under different loading conditions with the accelerated life experiment under the same loading conditions, and obtain the SIS logic controller life error Y.

[0047] After obtaining the life error, the process proceeds to step S130 . Step S130 establishes a data-driven model that characterizes the relationship between the combination factor input parameter and the life error parameter, and solves the current data-driven model based on each combination factor sample and the corresponding life error to obtain the optimal combination factor data set.

[0048] In one embodiment, Figure 4 As shown, in step S130 of obtaining the optimal combination factor data group, it includes: the first step, according to each combination factor sample and the corresponding life error, the data driven model is fitted to obtain the initial relationship model; the second step, based on the mixed point criterion, a new input sample point is obtained, and the initial relationship model is used to obtain the response value of the current new sample point; the third step, according to the new sample point and the corresponding response value, using the preset convergence judgment condition, it is judged whether the current relationship model meets the application requirements. If it does, the numerical range of each factor corresponding to the minimum life value of the current relationship model is used as the optimal combination factor data group.

[0049] In addition, if the current relationship model does not meet the application requirements, new sample points are added based on the point-adding criteria and the corresponding response values ​​are obtained using the current initial relationship model, thereby returning to the first step to update the initial relationship model based on the samples of each combined factor and the corresponding life error, as well as the newly added sample points and their response values.

[0050] In one embodiment, the process of solving the data-driven model includes: continuously adjusting the factor data in the initial sample space to obtain the minimum value of the life error parameter output by the model, thereby obtaining the optimal combination factor data group.

[0051] Additionally, the data-driven model is implemented using the Kriging surrogate model.

[0052] Specifically, according to the combined factor samples and the corresponding life error results obtained in step S120, an initial relationship model of the data-driven model of the relationship between the combined factor samples and the life error is established. Then, based on the mixed point-adding criterion, the initial relationship model is used to obtain a new life error result according to the new input sample point. Finally, the new sample points are continuously adjusted to obtain the minimum value of the life error. If the new sample points and the life error results meet the preset convergence judgment conditions, the numerical range of each factor of the sample point is the optimal combined factor data group. If the new input sample point does not meet the convergence judgment condition, it is necessary to add new sample points based on the point-adding criterion, and then obtain a new life error result according to the initial relationship model. Finally, the initial relationship model is updated using the obtained data until the sample points and the corresponding life error results meet the convergence judgment conditions.

[0053] After obtaining the optimal combination factor data group, the process proceeds to step S140. Step S140 obtains the second life data of the optimal combination factor data group under different input current conditions, thereby constructing a life evaluation model suitable for the SIS logic controller based on the basic failure physical model and using different input current data and the corresponding second life data.

[0054] In one embodiment, the specific steps of the life evaluation model construction process are: first, based on the three-dimensional finite element model of the SIS logic controller constructed in step S120, the optimal combination factor data group is used as the controller operating environment condition; then, multiple groups of input current data are applied to the controller in the finite element model to obtain the second life data under different input current sample conditions; finally, based on the basic failure physics model, according to different input current samples and the corresponding second life data, a life evaluation model that characterizes the relationship between current and life is obtained.

[0055] In one embodiment, in the process of constructing the life evaluation model, it includes: first, according to the input current sample and the corresponding second life data sample, according to the preset ratio of the training data volume to the verification data volume, a training data group and a verification data group are obtained; then, according to the constructed training data group, the above-mentioned basic failure physical model is fitted to obtain an initial life prediction model containing an initial fitting coefficient; finally, according to the constructed verification data group, the initial life prediction model is verified to update the initial fitting coefficient to obtain a life evaluation model containing an updated fitting coefficient. Among them, the ratio of the data volume in the training data group to the data volume in the verification data group is 4:1.

[0056] The above basic failure physical model is expressed by the following expression:

[0057]

[0058] Among them, t represents the life parameter, E represents the failure acceleration energy, k represents the Boltzmann constant, i represents the current parameter, A, They represent the coefficients to be fitted.

[0059] Specifically, the optimal combination factor data group obtained in step S130 is used as the environmental condition for the operation of the controller. Based on this environment, multiple groups of input current data are applied to the controller in the finite element model to obtain the corresponding second life data. As for the input current data obtained and the corresponding second life data obtained, they are divided into a training data group and a verification data group according to a preset ratio of 4:1. The training data group accounting for 4 / 5 of the data is used to fit the basic failure physics model to obtain an initial life prediction model containing initial fitting coefficients. The verification data group accounting for 1 / 5 of the data is used to verify the initial life prediction model, update the initial fitting coefficients, and obtain a life evaluation model containing updated fitting coefficients.

[0060] After the life evaluation model is obtained, the process proceeds to step S150 . Step S150 obtains a dynamic life evaluation result using the life evaluation model according to the real-time input current data of the SIS logic controller to be evaluated.

[0061] Specifically, through the life evaluation model, dynamic life evaluation results can be obtained by inputting current data, thereby providing guidance for actual work and reliability analysis, and forming the integrity assessment capability of the SIS logic controller.

[0062] Embodiment 2

[0063] Based on the above-mentioned first embodiment, the embodiment of the present invention provides a specific embodiment of a method for evaluating the life of a SIS logic controller.

[0064] S1 determines the initial sample space X according to the field working conditions of the petrochemical industry, taking into full account the external condition variable characteristics that affect the SIS logic controller and the numerical range of each characteristic parameter.

[0065] Furthermore, in S1, according to the fault tree, the main external influencing factors affecting the SIS logic controller in n petrochemical conditions are listed, and the space X is filled based on the optimal Latin hypercube criterion, and the initial sample space X = [x 1 ,x 2 ,x 3 ,…,x n ] T .

[0066] S2 loads the finite element model in sequence according to the initial boundary condition sample constructed by Latin hypercube. Rated current is applied to the board chip, and multi-physics field simulation is carried out by comsol fatigue module to obtain the finite element simulation life result of SIS logic controller. And compare the result with the accelerated life experiment under the same conditions to obtain the service life error y of SIS logic controller. Among them, according to the characteristics of electronic components, their failure is mostly due to the accumulation of Joule heat under the electro-thermal effect after the chip is powered on for a long time, resulting in chip breakdown.

[0067] Further, in S2, S21: construct a three-dimensional finite element model of the SIS logic controller in the comsol software, load the rated operating current to the chip of the board integrated circuit, and complete the material property setting, meshing and other processes. According to the boundary condition characteristic parameter sample constructed by the Latin hypercube, load the boundary conditions in sequence, and in the fatigue module, combine the linear cumulative damage theory to obtain the SIS logic finite element simulation life result y FEM .

[0068] Further, in S2, S22: through the accelerated aging test platform, an accelerated aging test is performed under the same boundary conditions and current conditions to obtain the experimental life y EXP , the experimental life y EXP Compare the life results obtained by finite element analysis to obtain the life error under each corresponding condition, that is, y = |y EXP -y FEM |.

[0069] S3 establishes a data-driven model between multi-boundary input parameters and the life error, takes the minimum error value as the optimization goal, uses the Kriging proxy model to construct the corresponding relationship between the input characteristic variables x and y, and optimizes the error y to obtain the optimal boundary conditions, updates the results back to the numerical simulation, obtains the optimal finite element model, and obtains the optimal boundary conditions.

[0070] Further, in S3, S31: construct variable X = [x 1 ,x 2 ,x 3 ,…,x n ] T The corresponding result Y={y (1) ,y (2) ,…,y (n)} T The corresponding relationship expression of , select the Kriging model expression as:

[0071] y(x)=f T (x)β+z(x)

[0072] Among them, β is the regression coefficient; f(x) is the regression function about x; the random term z(x) is a random function with a mean of zero and its covariance is expressed as:

[0073]

[0074] Among them, R(θ,x i ,x j ) represents any two sample points x i and x j The spatial correlation degree of ; θ is the model parameter to be trained, σ Z Represents the variance of the random function z(x). The correlation between two points in the Kriging surrogate model is related to the distance between the two points. The correlation function obtained using the exponential form is expressed as:

[0075]

[0076] Where n is the number of variables, x i k and x j k is the sample point x i and x j The kth component of k Indicates the model parameters to be trained under the specified component sample points.

[0077] Further, in S3, S32: a genetic algorithm is used to optimize the agent model, and the optimization model is shown in the following formula to achieve the optimization result with the minimum y value:

[0078]

[0079] Among them, x 1-min 、x 1-max Represents the variable x 1 The minimum and maximum values ​​of x 2-min 、x 2-max Represents the variable x 2 The minimum and maximum values ​​of x n-min 、x n-max Represents the variable x n In step S32, by searching for the minimum y value, the best optimization result value of each variable is obtained, thereby forming an optimal combination factor data set.

[0080] Further, in S3, S33: after obtaining the optimal x boundary value, the value is updated back to the finite element model, so as to further calculate the life of the SIS logic controller according to the optimal finite element model.

[0081] S4 is based on the finite element numerical model with optimal boundary conditions. Different input currents within the safe current range are loaded to obtain a large amount of life result data of SIS logic controllers, that is, to obtain the second life data corresponding to different input current samples. Combined with the typical Arrhenius failure physics method of electronic devices, a failure physics model suitable for SIS logic controllers is constructed.

[0082] Further, in S4, S41: according to the safe current range in the manual, a current range (i±25%i) is applied to the board chip to obtain the life result t of group q.

[0083] Further, in S4, S42: According to the failure physics equation, the characteristic life t of the reliability life of electronic products is usually represented by the highest chip junction temperature on the board. Since the junction temperature is positively correlated with the current value, the life t of this patent is represented by the rated current i:

[0084]

[0085] Where E is the failure acceleration energy, and k is the Boltzmann constant. Substitute t and i obtained in step S41 into the above formula, and fit into the established life evaluation model. Through the established life evaluation model and life data, the coefficients A and -E of the failure physical equation can be further obtained. a / k.

[0086] Embodiment 3

[0087] Based on the above-mentioned Embodiment 1 and Embodiment 2, the embodiment of the present invention further provides another specific embodiment of a method for evaluating the life of a SIS logic controller.

[0088] Step T1: Based on the field working conditions of the petrochemical industry, the characteristics of the external condition variables affecting the SIS logic controller and the numerical range of each characteristic parameter are fully considered, and the Latin hypercube sampling method is used to design preliminary experimental variables and determine the initial sample space X.

[0089] In this embodiment, Figure 2 The present invention is a flowchart of the step of determining an initial sample space in the method for evaluating the life of a SIS logic controller according to an embodiment of the present application.

[0090] According to the characteristic variables of the operating conditions influencing factors of SIS logic controllers in the petrochemical background, the external input variable condition library X and the numerical range of each condition that affect the reliability of SIS logic control are established, and the ambient temperature T, board spacing d, and vibration amplitude A are taken as external input variables. Among them, the range of input variables is shown in Table 1:

[0091] Table 1 Input feature variable range

[0092]

[0093] Filling the space X=[x based on the optimal Latin hypercube criterion T ,x A ,x d ] T , 20 initial samples of 3 variables are constructed through normalization. Figure 3 This is an example diagram of the spatial distribution of initial sample points in the method for evaluating the life of a SIS logic controller according to an embodiment of the present application. The three-dimensional spatial distribution constructed based on the above 3 variables and 20 initial samples is as follows: Figure 3 shown.

[0094] Combined with the characteristic parameter range of the boundary conditions in Table 1, the initial sample space is denormalized to obtain a partial experimental table as shown in Table 2.

[0095] Table 2 Input features and output result sample table

[0096]

[0097] Then, enter step T2, and load the finite element model in sequence according to the initial boundary condition sample constructed by the Latin hypercube. Apply the rated current to the board chip, perform multi-physics field simulation using the comsol fatigue module, and take the chip exceeding the fatigue limit as the failure basis to obtain the SIS logic controller finite element simulation life results. According to the characteristics of electronic components, the cause of failure is mostly due to the accumulation of Joule heat under the electro-thermal effect after the chip has been powered on for a long time, resulting in chip breakdown.

[0098] On the other hand, building an accelerated aging test platform can provide external test variables included in X to meet the test range. The simulation results are compared with the accelerated life test under the same conditions to obtain the service life error Y of the SIS logic controller.

[0099] In this embodiment, Figure 4 The present invention is a flow chart of the step of optimizing the boundary conditions of the finite element model in the method for evaluating the life of the SIS logic controller according to an embodiment of the present application.

[0100] like Figure 4 As shown in the figure, after determining the initial sample space, the actual life of the SIS logic controller obtained by the experiment is compared with the life of the SIS logic controller calculated by the finite element numerical simulation, and the life error Y of the SIS logic controller is obtained. Among them, the steps for the life results obtained by the experiment are: first, build an accelerated aging test platform, provide external test variables contained in X, and meet the test range. Then, the accelerated aging experiment is carried out using the same boundary conditions as the finite element numerical simulation. Finally, the actual life of the SIS logic controller under the accelerated aging experiment is obtained.

[0101] The steps for calculating the service life by finite element numerical simulation are as follows: first, the boundary conditions are set according to the X range, and a finite element model with three-dimensional geometric model, material property setting, and meshing is established. Then, according to the thermal stress failure physical model, multi-physics field joint simulation is performed. Finally, the service life of the SIS logic controller is calculated.

[0102] In the embodiment of the present invention, the SIS logic controller three-dimensional finite element model is constructed using the comsol software, and the rated working current of 5A is loaded on the chip of the board integrated circuit to complete the processes such as material property setting, mesh division, and heat dissipation coefficient setting. The boundary condition samples shown in Table 2 are loaded accordingly. In the fatigue module, combined with the linear cumulative damage theory, the chip temperature exceeding the rated temperature by 75% is used as the experimental cutoff basis, and the SIS logic finite element simulation life result y is obtained accordingly. FEM Through the accelerated aging test platform, the accelerated aging test is carried out under the same boundary conditions and current conditions to obtain the experimental life y EXP , calculate the difference between the two results y=|y EXP -y FEM |. As shown in Table 2.

[0103] After obtaining the life error of the SIS logic controller, enter step T3 to establish a data-driven model between the multi-boundary input parameters and the life error. Taking the minimum error value as the optimization goal, the Kriging proxy model is used to construct the corresponding relationship between the input characteristic variables X and Y, and optimize the error Y to obtain the optimal boundary conditions. In the solution, the mixed point addition criterion is combined to make the proxy model meet the convergence requirements or the number of points added reaches the upper limit, and the optimization result of X when the Y value is minimized is output. The results are updated back to the numerical simulation to obtain the optimal finite element model.

[0104] like Figure 4 As shown in the figure, after obtaining the service life error of the SIS logic controller, an error compensation proxy model is constructed to optimize the boundary conditions of the finite element model. The steps for constructing the error compensation proxy model are: first prepare the initial data and the corresponding response value, then select the modeling method, construct the Kriging proxy model to establish the mapping relationship between the error Y and the input condition X, obtain the update point based on the mixed point addition criterion, and calculate the response value of the update point. According to the new sample points and the corresponding response values, it is judged whether the current relationship model meets the convergence requirements or whether the number of added points reaches the upper limit. If the conditions are met, the result of X corresponding to the current minimum Y value is output, and the output X value is used to update the boundary conditions set in the finite element numerical simulation. If the conditions are not met, the sample points are added based on the point addition criterion and returned to the previous step, and the response values ​​of the sample points and the newly added sample points are added to the sample set to update the initial proxy model.

[0105] In this embodiment, an external input variable condition library X=[x T ,x A ,x d ] T The corresponding result Y={y 1 ,y 2 ,…,y 20 The corresponding expression of}, based on the Kriging model, the minimum y value optimization expression is:

[0106]

[0107] The genetic algorithm is used to optimize the agent model, and the optimal result to achieve the minimum y value is [x T =38,x A =29.3,x d =66.7], y = 30.56. The above optimal X boundary condition characteristic parameters are updated back to the finite element model.

[0108] Next, enter step T4, based on the finite element numerical model with optimal boundary conditions, load the current i within the safe current range to obtain multiple sets of life results. Combined with the typical Arrhenius failure physics method of electronic devices, the failure physics equation of the SIS logic controller is constructed. 4 / 5 groups of life data are selected to fit the coefficients of the failure physics equation, and the remaining 1 / 5 data is used as the model verification part, and the coefficients are revised.

[0109] In an embodiment of the present invention, Figure 5 This is a flow chart of constructing a SIS logic controller failure physical equation in the method for evaluating the life of a SIS logic controller in an embodiment of the present application. Figure 5 As shown in the figure, the steps to construct the SIS logic controller failure physical equation are as follows: first, under the optimal boundary conditions, apply a current i within the rated range to the chip to obtain q groups of life sample results. Then, the SIS logic controller failure physical equation is constructed, and the unknown coefficients of the equation are verified and revised using the data results of the 80% group, and the data results of the remaining 20% ​​group are used to verify the equation coefficients.

[0110] Specifically, in step T41, under the optimal boundary conditions, the current range i = [3.5A, 6.25A] is applied to the board chip, and other parameter settings remain unchanged, and 100 sets of life results y are obtained. FEM To distinguish from the results in Table 2, the life results under the optimal boundary conditions are represented by t, and the result is t = [240.5, 240, ..., 243].

[0111] In step T42, according to the failure physics equation, the life t and the rated current i are expressed as:

[0112]

[0113] The coefficients A and -E in the characteristic life model are adjusted using the first 80 sets of data obtained in step S41. a / k is fitted to obtain the initial coefficients A = 8002 and -E a / k=0.12. Substituting into the above formula, the fitting result is verified by using the last 20 sets of life data, and further corrected to obtain the final result A=8001.55 and -E a / k=0.122.

[0114] Embodiment 4

[0115] Based on the above-mentioned Embodiments 1 to 3, the present invention provides a computer-readable storage medium, on which a computer program is stored, and the computer program is executed to run a series of instructions of a method step for evaluating the life of a SIS logic controller. The computer program can run computer instructions, and the computer instructions include computer program code, which can be in source code form, object code form, executable file or some intermediate form, etc.

[0116] Computer-readable storage media may include: any entity or device that can carry computer program code, recording media, USB flash drives, mobile hard disks, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0117] It should be noted that the content contained in computer-readable storage media can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, the content contained in computer-readable storage media does not include electrical carrier signals and telecommunication signals.

[0118] Embodiment 5

[0119] Based on the above-mentioned Embodiment 1 to Embodiment 3, the present invention also provides a system for evaluating a SIS logic controller. The system is used to implement the above-mentioned method for evaluating the life of a SIS logic controller. Figure 6 Schematic diagram of the overall structure of the system for evaluating the life of a SIS logic controller according to an embodiment of the present application. Figure 6As shown, the system includes: an influencing factor sample space generation module 61, a life error sample data generation module 62, an optimal combination factor generation module 63, a life evaluation model generation module 64 and an online evaluation module 65.

[0120] Specifically, the influencing factor sample space generation module configuration 61 is implemented according to the method described in the above step S110, and is configured to generate an initial sample space according to the factors affecting the working condition of the SIS logic controller; the life error sample data generation module 62 is implemented according to the method described in the above step S120, and is configured to load the samples under each combination of factors in the initial sample space on a preset finite element model to obtain the first life result under the corresponding combination of factors, and then compare the first life result with the result of the accelerated life test under the same conditions to obtain the life error; the optimal combination factor generation module 63 is implemented according to the method described in the above step S130, and is configured to establish a relationship between the input parameters characterizing the combination factors and the life error parameters. The data-driven model of the relationship between them is constructed, and according to each combination factor sample and the corresponding life error, the data-driven model is solved to obtain the optimal combination factor data group; the life evaluation model generation module 64 is implemented according to the method described in the above step S140, and is configured to obtain the second life data of the optimal combination factor data group under different input current conditions, so as to construct a life evaluation model suitable for the SIS logic controller based on the basic failure physical model and using different input current data and the corresponding second life data; the online evaluation module 65 is implemented according to the method described in the above step S150, and is configured to obtain a dynamic life evaluation result by using the life evaluation model according to the real-time input current data of the SIS logic controller to be evaluated.

[0121] The present invention discloses a method and system for evaluating the life of a SIS logic controller. The method and system take into account the operating factors under the background of a petrochemical environment, such as board spacing, environmental humidity, vibration and other factors, and add these boundary condition features when establishing a finite element model, so as to more realistically reflect the life of the SIS logic controller under real operating and environmental conditions; at the same time, in the process of setting the boundary condition parameters, the optimal Latin hypercube is used to construct the sample space to improve the spatial filling degree of the sample. In addition, the present invention organically combines finite element numerical simulation and accelerated aging experiment, calculates the error between the finite element model and the experiment under corresponding conditions, and constructs the relationship between the input boundary condition characteristic parameter and the error, and then uses the Kriging proxy model to optimize the error; updates the boundary conditions of the finite element model to obtain the optimal finite model close to the experiment; adopts the optimal finite element model, which can greatly reduce the number of experimental samples, reduce manpower and financial resources, and at the same time ensure the number of samples, and improve the accuracy of subsequent life model establishment. In addition, the present invention also compares a variety of accelerated life models, compares the correlation coefficient of the life model and the mean error RMSE and other indicators, to establish an accelerated life model suitable for the SIS logic controller, which has certain guidance for the actual situation.

[0122] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any person familiar with the technology within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0123] In the description of the present invention, unless otherwise specified, "plurality" means two or more than two; the orientations or positional relationships indicated by the terms "upper", "lower", "left", "right", "inner", "outer", "front end", "rear end", "head", "tail", etc. are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0124] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "connected" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0125] It should be understood that the embodiments disclosed in the present invention are not limited to the specific structures, processing steps or materials disclosed herein, but should be extended to equivalent substitutions of these features understood by ordinary technicians in the relevant field. It should also be understood that the terms used herein are only used for the purpose of describing specific embodiments and are not meant to be limiting.

[0126] The "one embodiment" or "embodiment" mentioned in the specification means that the specific features, structures or characteristics described in conjunction with the embodiment are included in at least one embodiment of the present invention. Therefore, the phrases "one embodiment" or "embodiment" appearing in various places throughout the specification do not necessarily refer to the same embodiment.

[0127] Although the embodiments disclosed in the present invention are as above, the contents described are only embodiments adopted for facilitating the understanding of the present invention and are not intended to limit the present invention. Any technician in the technical field to which the present invention belongs can make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in the present invention, but the patent protection scope of the present invention shall still be subject to the scope defined in the attached claims.

Claims

1. A method for evaluating the life of a SIS logic controller, characterized in that: include: Step 1: Generate an initial sample space based on various factors that affect the operating conditions of the SIS logic controller; Step 2: Load the samples under each combination factor in the initial sample space onto a preset finite element model to obtain a first life result under the corresponding combination factor, and then compare the first life result with the result of the accelerated life test under the same conditions to obtain a life error; Step 3: Establish a data-driven model that characterizes the relationship between the combination factor input parameters and the life error parameters, and solve the data-driven model according to each combination factor sample and the corresponding life error to obtain the optimal combination factor data set; Step 4: obtaining the second life data of the optimal combination factor data group under different input current conditions, thereby constructing a life evaluation model suitable for the SIS logic controller based on the basic failure physical model and using different input current data and the corresponding second life data; Step 5: According to the real-time input current data of the SIS logic controller to be evaluated, the life evaluation model is used to obtain a dynamic life evaluation result.

2. The method according to claim 1, characterized in that The step one comprises: Determine various variable factors and corresponding parameter ranges; According to the parameter range of each variable factor, the optimal Latin hypercube criterion is used to fill the sample space to form a three-dimensional sample space distribution; The three-dimensional sample space distribution is normalized to obtain the initial sample space.

3. The method according to claim 1 or 2, characterized in that: The process of generating the first life result includes: A three-dimensional finite element model of the SIS logic controller is constructed, and the rated current is loaded on the current controller. Then, the samples under each combination factor in the initial sample space are loaded on the three-dimensional finite element model in sequence. Combined with the linear cumulative damage theory, the controller chip temperature exceeding the preset threshold temperature is used as the cutoff basis to obtain the first life results under different sample loading conditions.

4. The method according to any one of claims 1 to 3, characterized in that The process of solving the data-driven model includes: continuously adjusting the factor data in the initial sample space so that the life error parameter output by the model reaches the minimum value, thereby obtaining the optimal combination factor data group.

5. The method according to claim 4, characterized in that The step three includes: The first step is to fit the data-driven model according to the samples of each combined factor and the corresponding life errors to obtain an initial relationship model; The second step is to obtain new input sample points based on the mixed point addition criterion, and obtain the response value of the current new sample point by using the initial relationship model; The third step is to determine whether the current relationship model meets the application requirements based on the new sample points and the corresponding response values ​​and using the preset convergence judgment conditions. If so, the numerical range of each factor corresponding to the minimum life value of the current relationship model is used as the optimal combination factor data group.

6. The method according to claim 5, characterized in that If the current relationship model does not meet the application requirements, new sample points are added based on the point adding criteria and the corresponding response values ​​are obtained using the current initial relationship model, thereby returning to the first step to update the initial relationship model based on the samples of each combination factor and the corresponding life error, as well as the newly added sample points and their response values.

7. The method according to any one of claims 4 to 6, characterized in that: The data driven model is implemented using the Kriging proxy model.

8. The method according to any one of claims 1 to 7, characterized in that: In the step 4, it includes: According to the three-dimensional finite element model of the SIS logic controller, the optimal combination factor data set is used as the controller operating environment condition; Apply multiple sets of input current data to the controller in the finite element model to obtain second life data under different input current sample conditions; Based on the basic failure physics model, according to different input current samples and the corresponding second life data, the life evaluation model characterizing the relationship between current and life is obtained.

9. The method according to claim 8, characterized in that The step of obtaining the life evaluation model characterizing the relationship between current and life based on the basic failure physical model according to different input current samples and corresponding second life data includes: According to the input current sample and the corresponding second life data sample, a training data group and a verification data group are obtained according to a preset ratio of the training data amount to the verification data amount, wherein the ratio of the training data amount to the verification data amount is 4:1; Fitting the basic failure physics model according to the training data set to obtain an initial life prediction model containing initial fitting coefficients; The initial life prediction model is verified according to the verification data set to update the initial fitting coefficients to obtain the life assessment model containing the updated fitting coefficients.

10. The method according to claim 8 or 9, characterized in that: The basic failure physics model is expressed using the following expression: Among them, t represents the life parameter, E represents the failure acceleration energy, k represents the Boltzmann constant, i represents the current parameter, A, They represent the coefficients to be fitted.

11. A computer-readable storage medium, characterized in that: It contains a series of instructions for executing the method steps as claimed in any one of claims 1 to 10.

12. A system for evaluating the life of a SIS logic controller, characterized in that: include: An influencing factor sample space generation module configured to generate an initial sample space according to various factors affecting the operating conditions of the SIS logic controller; A life error sample data generation module, configured to load samples under each combination factor in the initial sample space onto a preset finite element model, obtain a first life result under the corresponding combination factor, and then compare the first life result with a result of an accelerated life test under the same conditions to obtain a life error; An optimal combination factor generation module is configured to establish a data-driven model that characterizes the relationship between the combination factor input parameters and the life error parameters, and solve the data-driven model according to each combination factor sample and the corresponding life error to obtain an optimal combination factor data set; A life evaluation model generation module, configured to obtain the second life data of the optimal combination factor data group under different input current conditions, so as to construct a life evaluation model suitable for the SIS logic controller based on the basic failure physical model and using different input current data and the corresponding second life data; The online evaluation module is configured to obtain a dynamic life evaluation result by using the life evaluation model according to the real-time input current data of the SIS logic controller to be evaluated.

Images