Simulation analysis method for fatigue life of engine oil nozzle matching part structure

Abstract

The invention is applicable to the technical field of engine oil nozzles, and particularly relates to an engine oil nozzle matching part structure fatigue life simulation analysis method, which comprises the following steps of: acquiring geometrical characteristics and motion characteristics of an oil nozzle; establishing a space incidence matrix of each component in the oil nozzle based on the geometrical characteristics to obtain position information; fusing the motion features and the position information to obtain fused features; predicting a crack propagation path of the matching part according to the fusion features; calculating a fatigue index based on the geometric features and the motion features, and generating a failure competition graph based on the fatigue index; and obtaining the fatigue life of the matching part according to the crack propagation path and the failure competition graph. Therefore, the engine fuel injection nozzle matching part structure fatigue life simulation analysis method provided by the invention can solve the problem that the fatigue life of the fuel injection nozzle matching part is difficult to accurately simulate and determine because the influence of multiple failure modes on the fatigue life is neglected in the prior art.

Description

Technical Field

[0001] This application belongs to the field of engine fuel injector technology, and in particular relates to a simulation analysis method for fatigue life of engine fuel injector assembly structure. Background Technology

[0002] In the existing technology, the fuel injector assembly is the core component of the diesel engine fuel injection system. It is usually composed of a precision-fitted needle valve and a needle valve body. Its function is to control the combustion process through the precise injection of high-pressure fuel, which directly affects the engine's efficiency and emission performance.

[0003] In existing technologies, fatigue life prediction of fuel injector components mainly relies on accelerated aging tests in a laboratory environment, combined with finite element simulation analysis of stress distribution, to assess the damage accumulation process of fuel injector components under high-frequency, high-pressure alternating loads. However, laboratory environments often cannot fully simulate the complex operating conditions of a real-world engine. Furthermore, existing technologies typically analyze fatigue fracture, wear, and corrosion of the fuel injector separately, neglecting their interactions, leading to conservative fatigue life predictions. Therefore, existing technologies overlook the impact of multiple failure modes on fatigue life, making it difficult to accurately simulate and determine the fatigue life of fuel injector components. Summary of the Invention

[0004] This application provides a method for simulating and analyzing the fatigue life of an engine fuel injector assembly, which can solve the problem of neglecting the influence of multiple failure modes on fatigue life, making it difficult to accurately simulate and determine the fatigue life of the fuel injector assembly.

[0005] In a first aspect, embodiments of this application provide a method for simulating and analyzing the fatigue life of an engine fuel injector assembly structure, including: The geometric and kinematic features of the fuel injector are obtained; wherein the geometric features include size, shape and surface roughness, and the kinematic features are used to reflect the deformation and vibration of the fuel injector under different operating conditions. Based on the geometric features, a spatial correlation matrix of each component in the fuel injector is established to obtain position information; wherein, the components include a coupler, a spring, an O-ring, and a filter screen, and the coupler includes a needle valve and a needle valve body; The motion features and the location information are fused to obtain fused features; The crack propagation path of the pair is predicted based on the fusion features; A fatigue index is calculated based on the geometric features and the motion features, and a failure competition map is generated based on the fatigue index; wherein, the failure competition map includes the probabilities of multiple failure modes and the corresponding failure time windows; The fatigue life of the component is obtained based on the crack propagation path and the failure competition diagram.

[0006] The technical solutions described in this application embodiment have at least the following technical effects: The fatigue life simulation analysis method for engine injector assembly structure provided in this application involves: acquiring the geometric and kinematic features of the injector; establishing a spatial correlation matrix for each component in the injector based on the geometric features to obtain positional information; fusing the kinematic and positional information to obtain fused features; predicting the crack propagation path of the assembly based on the fused features; calculating the fatigue index based on the geometric and kinematic features; generating a failure competition diagram based on the fatigue index; and obtaining the fatigue life of the assembly based on the crack propagation path and the failure competition diagram. Therefore, the fatigue life simulation analysis method for engine injector assembly structure provided in this application constructs a spatial correlation matrix, spatially aligns and dynamically correlates geometric and kinematic features to form fused features. Utilizing the fused features to construct a dynamic crack propagation model, generating a failure competition diagram through fatigue index calculation, and comprehensively evaluating fatigue life by combining the crack propagation path and the failure competition diagram, this method facilitates adaptive prediction under operating conditions and improves the practicality and reliability of injector assembly life prediction.

[0007] In one possible implementation of the first aspect, predicting the crack propagation path of the pair based on the fused features includes: A fatigue feature vector is obtained based on the fusion features; wherein, the fatigue feature vector is used to quantify the fatigue sensitivity of the component under cyclic loading; The attention weights for the fatigue life of the component are obtained based on the fatigue feature vector; wherein the attention weights are used to reflect the degree of contribution of each component in the fatigue feature vector to the prediction of the crack propagation path. The transformation feature vector is determined based on the attention weight; wherein the transformation feature vector is used to reflect the degree of fatigue influence between the needle valve and the needle valve body in the coupling. The crack propagation path is obtained by predicting based on the fatigue feature vector and the transformed feature vector.

[0008] In one possible implementation of the first aspect, obtaining the fatigue feature vector based on the fused features includes: Based on the fusion features of the device, a first correlation between the needle valve and the needle valve body in the device is calculated to obtain a first correlation vector; wherein, the first correlation vector is used to reflect the linear correlation between the needle valve and the needle valve body in the fusion feature space of the device. Based on the first correlation vector and the fusion feature, the second correlation between the needle valve and the needle valve body in the pair is calculated to obtain the second correlation vector; wherein, the second correlation vector is used to reflect the nonlinear correlation between the needle valve and the needle valve body in the fusion feature space of the pair; The fatigue feature vector of the device is obtained by fusing the first correlation vector with the second correlation vector.

[0009] In one possible implementation of the first aspect, determining the transformed feature vector based on the attention weights includes: The attention feature vectors of the needle valve and the needle valve body in the device are determined based on the attention weights. The transformation feature vectors of the needle valve and the needle valve body in the device are obtained based on the attention feature vector.

[0010] In one possible implementation of the first aspect, determining the attention feature vectors of the needle valve and the needle valve body in the coupler based on the attention weights includes: The initial feature vector of the device is obtained based on the motion characteristics; The initial feature vector is updated according to the attention weights to obtain the attention feature vector of the device.

[0011] In one possible implementation of the first aspect, the step of predicting the crack propagation path based on the fatigue feature vector and the transformed feature vector includes: The vertices of the meshed component are defined as nodes; For each node, the fatigue feature vectors of the neighboring nodes are aggregated, and the crack propagation probability of each node is predicted by combining the transformed feature vectors. The crack propagation path is obtained based on the crack propagation probability.

[0012] In one possible implementation of the first aspect, the step of calculating a fatigue index based on the geometric features and the motion features, and generating a failure competition map based on the fatigue index, includes: Construct a correlation network between the geometric features and the motion features; wherein the correlation network is used to analyze the correlation and degree of influence between the components. The fatigue index of each component is calculated based on the associated network; The probability of each failure mode is calculated based on the fatigue index, and the corresponding failure time window is predicted.

[0013] In one possible implementation of the first aspect, acquiring the geometric and kinematic features of the fuel injector includes: Through simulation analysis, the motion state of the fuel injector under different operating conditions is simulated, and the motion characteristics of each component are extracted.

[0014] In one possible implementation of the first aspect, obtaining the fatigue life of the mating component based on the crack propagation path and the failure competition map includes: The crack propagation path is converted into a time series to obtain the crack propagation time; The fatigue life is obtained by correcting the crack propagation time based on the failure competition diagram.

[0015] In one possible implementation of the first aspect, the method further includes: The structural parameters of the fuel injector are determined within a preset range based on the fatigue life.

[0016] Secondly, embodiments of this application provide a fatigue life simulation analysis device for engine fuel injector assembly structures, comprising: The acquisition module is used to acquire the geometric and kinematic features of the fuel injector; wherein, the geometric features include size, shape and surface roughness, and the kinematic features are used to reflect the deformation and vibration of the fuel injector under different operating conditions; The position information module is used to establish a spatial correlation matrix of each component in the fuel injector based on the geometric features to obtain position information; wherein, the components include a coupler, a spring, an O-ring, and a filter screen, and the coupler includes a needle valve and a needle valve body; A feature fusion module is used to fuse the motion features and the position information to obtain fused features; A crack propagation path module is used to predict the crack propagation path of the pair based on the fusion features. The failure competition graph module is used to calculate the fatigue index based on the geometric features and the motion features, and generate a failure competition graph based on the fatigue index; wherein, the failure competition graph includes the probabilities of multiple failure modes and the corresponding failure time windows; The fatigue life module is used to obtain the fatigue life of the component based on the crack propagation path and the failure competition diagram.

[0017] Thirdly, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method as described in any one of the first aspects above.

[0018] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in any of the first aspects above.

[0019] Fifthly, embodiments of this application provide a computer program product that, when run on an electronic device, causes the electronic device to perform the method described in any one of the first aspects above.

[0020] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart illustrating the fatigue life simulation analysis method for an engine fuel injector assembly structure provided in one embodiment of this application. Figure 2 This is a schematic diagram of the implementation process of steps S100, S400, S410, S430, S431, S440, S500 and S600 in the fatigue life simulation analysis method for engine injector assembly structure provided in an embodiment of this application. Figure 3 This is a schematic diagram of crack length in the fatigue life simulation analysis method for engine fuel injector assembly structure provided in an embodiment of this application; Figure 4 This is a schematic diagram of the fatigue life in the fatigue life simulation analysis method for engine fuel injector assembly structure provided in an embodiment of this application; Figure 5 This is a schematic diagram of the fatigue life simulation analysis device for engine fuel injector assembly provided in this application embodiment; Figure 6 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0023] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0024] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0025] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0026] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0027] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0028] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0029] In related technologies, fatigue life prediction of fuel injector components mainly relies on accelerated aging tests in a laboratory environment, combined with finite element simulation analysis of stress distribution, to assess the damage accumulation process of fuel injector components under high-frequency, high-pressure alternating loads. However, laboratory environments often cannot fully simulate the complex operating conditions of a real vehicle engine. Furthermore, existing technologies typically analyze fatigue fracture, wear, and corrosion of the fuel injector separately, neglecting their interactions, leading to conservative fatigue life predictions. Therefore, existing technologies ignore the impact of multiple failure modes on fatigue life, making it difficult to accurately simulate and determine the fatigue life of fuel injector components.

[0030] To address the aforementioned issues, this application provides a method for simulating and analyzing the fatigue life of an engine injector assembly. This method involves acquiring the geometric and kinematic features of the injector; establishing a spatial correlation matrix for each component within the injector based on the geometric features to obtain positional information; fusing the kinematic and positional information to obtain fused features; predicting the crack propagation path of the assembly based on the fused features; calculating the fatigue index based on the geometric and kinematic features; generating a failure competition diagram based on the fatigue index; and obtaining the fatigue life of the assembly based on the crack propagation path and the failure competition diagram. Therefore, the engine injector assembly fatigue life simulation and analysis method provided in this application constructs a spatial correlation matrix, spatially aligning and dynamically associating geometric and kinematic features to form fused features. By utilizing the fused features to construct a dynamic crack propagation model, generating a failure competition diagram through fatigue index calculation, and comprehensively evaluating fatigue life by combining the crack propagation path and the failure competition diagram, this method facilitates adaptive prediction under operating conditions and improves the practicality and reliability of injector assembly life prediction.

[0031] The fatigue life simulation analysis method for engine injector assembly structure provided in this application embodiment can be applied to electronic devices. In this case, the electronic device is the executing subject of the fatigue life simulation analysis method for engine injector assembly structure provided in this application embodiment. This application embodiment does not impose any restrictions on the specific type of electronic device.

[0032] For example, electronic devices can be industrial computers, programmable logic controllers, embedded control systems, distributed control systems, tablet computers, laptops, ultra-mobile personal computers (UMPCs), netbooks, desktop computers, laptops, handheld computing devices, etc., but are not limited to these.

[0033] To better understand the fatigue life simulation analysis method for engine injector assembly structure provided in this application embodiment, the specific implementation process of the fatigue life simulation analysis method for engine injector assembly structure provided in this application embodiment will be described by way of example below.

[0034] Figure 1 This paper presents a schematic flowchart illustrating the fatigue life simulation analysis method for engine fuel injector assembly structures provided in an embodiment of this application. The fatigue life simulation analysis method for engine fuel injector assembly structures includes: S100 acquires the geometric and kinematic characteristics of the fuel injector. The geometric characteristics include size, shape, and surface roughness, while the kinematic characteristics reflect the deformation and vibration of the fuel injector under different operating conditions.

[0035] It is understandable that motion characteristics include stress distribution, deformation and / or vibration frequency.

[0036] For example, laser frequency comb 3D profile measurement technology can be used to obtain the size and shape of the fuel injector through interference signal spectrum analysis. Alternatively, the size and shape of the fuel injector can be obtained by using fiber optic probe (WFP) combined with CT scanning to acquire three-dimensional point cloud data. The surface roughness of the fuel injector can be obtained using white light interferometer (WLI). The three-dimensional model of the fuel injector obtained by CT scanning or laser measurement can be imported into ANSYS, fuel pressure pulses can be set, the bottom of the needle valve body can be fixed, and a preload can be applied to the contact surface between the needle valve and the spring. Finite element analysis can be used to simulate the stress distribution, deformation, and vibration frequency of the fuel injector under different operating conditions (i.e., different fuel pressure pulses) to obtain its motion characteristics.

[0037] S200: Based on geometric features, a spatial correlation matrix is ​​established for each component in the fuel injector to obtain positional information. The components include a coupling, spring, O-ring, and filter; the coupling includes a needle valve and a needle valve body.

[0038] For example, the topological relationships of the fuel injector assembly can be determined by constructing an association matrix between components based on graph theory. A 3D model is then built using CAD software based on the topological relationships and geometric features. Position information (including component name, coordinates, rotation angle, etc.) is obtained by determining the relative positions of the components through constraint relationships (coaxial constraints, distance constraints, and symmetry constraints, etc.) and mesh generation.

[0039] S300 fuses motion features and location information to obtain fused features.

[0040] For example, a Kalman filter algorithm can be used to fuse the motion characteristics and position information of the fuel injector, and the errors in the position information (such as displacement deviations caused by vibration) can be corrected based on the motion characteristics to obtain the fused features. Alternatively, the motion characteristics and position information can be concatenated to obtain the fused features.

[0041] S400, the crack propagation path of the pair is predicted based on the fusion features.

[0042] For example, the crack propagation path can be simulated using the extended finite element method (XFEM). Stress concentration regions are determined based on the stress distribution in the fusion characteristics. The initial crack is then identified based on these stress concentration regions and defined using the level set method. Crack tip singularities are separated using XFEM, and the stress intensity factors for Type I (opening) and Type II (slipping) cracks are calculated. The crack propagation direction is then calculated using the maximum circumferential stress criterion (STTMAX) based on the stress intensity factors. The crack propagation rate is predicted using the Paris formula (da / dN = c(ΔK)^n), where a is the crack depth or width, N is the number of stress cycles, ΔK is the range of stress intensity factors, and c and n are material constants. The crack propagation path of the component is obtained based on the crack propagation direction and crack propagation rate.

[0043] For example, the fusion features and historical crack data can be combined to use an LSTM network to learn the temporal dependence of crack propagation. The inputs are the fusion features and the initial crack, and the outputs are the crack propagation path (coordinate sequence) and the number of load cycles.

[0044] The S500 calculates the fatigue index based on geometric and kinematic features, and generates a failure competition map based on the fatigue index. The failure competition map includes the probabilities of various failure modes and their corresponding failure time windows.

[0045] It is understandable that failure competition maps are used to reflect the competitive relationships between different failure modes (such as fatigue fracture, wear, and corrosion).

[0046] For example, the probability of failure modes can be statistically analyzed using a principal-secondary plot (Pareto plot) based on historical failure data or accelerated life test (ALT) results, and a Weibull distribution can be fitted to the time data of each failure mode to obtain the failure time window, such as... Figure 3 As shown.

[0047] S600, the fatigue life of the component is obtained based on the crack propagation path and failure competition diagram.

[0048] For example, the number of cycles N required for a crack to propagate from an initial crack (such as a stress concentration region) to a critical crack (such as a crack extending to a stress concentration region at the needle valve head or a crack penetrating the needle valve cone) can be determined by the crack propagation path. fatigue The probability P of fatigue fracture is obtained from the Pareto diagram. fatigue Other failure mode probabilities P other and the time window T for each failure mode other The competing failure correction factor α (to reflect the reduction in fatigue life caused by other failure modes) is calculated based on the probability of fatigue fracture, the probabilities of other failure modes, and the time windows of each failure mode. The fatigue life of the component is then calculated based on the number of cycles required for the initial crack to propagate to the critical crack and the competing failure correction factor. For example, α = 1− other I(N fatigue ∈T other ), where I is the indicator function, N fatigue ∈T other The value is 1 if the fatigue life is 1, otherwise it is 0. The fatigue life of the component can be represented by the number of cycles, i.e., N = α. N fatigue .

[0049] In one possible implementation, please refer to Figure 2 S400, based on the fusion feature prediction, the crack propagation path of the component is obtained, including: S410, obtain the fatigue feature vector based on the fused features. The fatigue feature vector is used to quantify the fatigue sensitivity of the component under cyclic loading.

[0050] For example, the average stress level and average deformation of the component can be determined based on the fusion characteristics, the dominant frequency component of the vibration signal can be extracted by Fast Fourier Transform (FFT), and the above parameters can be combined into a vector form to obtain the fatigue feature vector.

[0051] S420: The attention weights for the fatigue life of the component are obtained based on the fatigue feature vector. These attention weights reflect the contribution of each component in the fatigue feature vector to the prediction of the crack propagation path.

[0052] For example, a fatigue feature vector can be input into a multilayer perceptron to generate a query vector, a key vector, and a value vector. An attention score can then be calculated based on the query vector Q, the key vector K, and the value vector V. For instance, the attention score Attention(Q,K,V) = softmax(QK). T / V, where, The dimension is vector. The attention score is determined as the attention weight for the fatigue life of the pair.

[0053] S430 determines the transformation feature vector based on attention weights. The transformation feature vector reflects the degree of fatigue influence between the needle valve and the needle valve body in the assembly.

[0054] For example, fatigue feature vectors of the needle valve and the needle valve body can be extracted separately, and the transformed feature vector can be obtained by weighted fusion based on attention weights.

[0055] S440, based on the fatigue feature vector and the transformed feature vector, predicts the crack propagation path.

[0056] For example, the corresponding fatigue feature vector and transformation feature vector can be calculated based on historical crack data, and a neural network (such as LSTM) can be trained to learn the temporal dependence of crack propagation. The input is the fatigue feature vector and transformation feature vector, and the output is the coordinates of the crack propagation path.

[0057] Through the above steps S410 to S440, the contribution of each feature to fatigue is clarified by weight visualization, the degree of fatigue influence between components is clarified, and the integration of physical model and data-driven technology is conducive to achieving high accuracy, high interpretability and high generalization (adapting to multiple working conditions) in the fatigue life prediction of fuel injector components, which is conducive to improving prediction efficiency and reducing test costs.

[0058] Optionally, please refer to Figure 2 S410, Based on the fusion features, the fatigue feature vector is obtained, including: S411, Based on the fusion features of the pair, calculate the first correlation between the needle valve and the needle valve body in the pair to obtain the first correlation vector. The first correlation vector reflects the linear correlation between the needle valve and the needle valve body in the fusion feature space.

[0059] For example, the linear correlation between the needle valve and the needle valve body can be quantified using the Pearson correlation coefficient matrix based on the fusion features of the components. The first correlation vector is obtained by assembling the correlation coefficients of each dimension in the fusion features into a vector.

[0060] S412, based on the first correlation vector and the fused features, calculate the second correlation between the needle valve and the needle valve body in the assembly, and obtain the second correlation vector. The second correlation vector reflects the nonlinear correlation between the needle valve and the needle valve body in the fused feature space.

[0061] For example, the fused features can be mapped to a high-dimensional space using kernel tricks, the mutual information value between the needle valve and the needle valve body can be calculated, and a second correlation vector can be obtained based on the mutual information value.

[0062] S413, the first correlation vector and the second correlation vector are fused to obtain the fatigue feature vector of the component.

[0063] For example, the weights of the first and second correlation vectors can be dynamically assigned using a Transformer model. A weighted concatenation and fusion strategy is then employed to combine the first and second correlation vectors to generate a fatigue feature vector.

[0064] Through steps S411 to S413, the linear coupling effect between the needle valve and its body in the fused feature space is comprehensively captured using a multidimensional correlation coefficient matrix. Nonlinear dependencies are quantified through mutual information, revealing hidden coupling mechanisms neglected by traditional methods. High-precision characterization of the fatigue features of the fuel injector assembly is achieved through step-by-step calculation and dynamic fusion of linear and nonlinear correlations. Compared to traditional methods, this reduces prediction errors and adapts to complex operating conditions. The fused vector contains multidimensional linear and nonlinear information, providing more comprehensive feature inputs for fatigue life prediction. Furthermore, dynamic weight allocation enhances model interpretability and robustness, ultimately contributing to high accuracy and high generalization in fatigue life prediction.

[0065] Optionally, please refer to Figure 2 S430, determines the transformed feature vector based on attention weights, including: S431, determine the attention feature vectors of the needle valve and the needle valve body in the pair according to the attention weight.

[0066] For example, the fusion features of the needle valve and the needle valve body can be concatenated into an input matrix, and a query, key, and value matrix can be generated through linear transformation. The attention score, i.e. the attention weight, can be calculated, and the attention weight and the value matrix can be weighted and summed to generate an attention feature vector.

[0067] S432, obtain the transformation feature vectors of the needle valve and the needle valve body in the pair based on the attention feature vector.

[0068] For example, residual connection and layer normalization can be used to convert the attention feature vector into a transformed feature vector. For instance, the attention feature vector is added to the input matrix in S431, retaining low-level information to obtain the residual result. The residual result is then normalized and further high-order features are extracted through a fully connected layer (FC) to obtain the transformed feature vector.

[0069] Through steps S431 to S432, attention weights highlight key features and reduce interference from irrelevant features. By jointly optimizing the attention mechanism and residual transformation, efficient extraction of fatigue features from the fuel injector assembly is achieved. Compared to traditional methods, this improves the accuracy of key feature recognition and reduces fatigue life prediction errors. The transformed feature vector possesses both discriminative and robust properties, providing high-quality input for subsequent predictions.

[0070] For example, please refer to Figure 2 S431, determine the attention feature vectors of the needle valve and needle valve body in the pair according to the attention weights, including: S4311, the initial feature vector of the device is obtained based on the motion features.

[0071] For example, time-domain features (such as mean, variance, peak value, etc.) and frequency-domain features (such as dominant frequency, amplitude, etc.) can be extracted from the motion characteristics (such as stress distribution, deformation and vibration frequency) of the components (needle valve and needle valve body), and then converted into the initial feature vector of the components.

[0072] S4312, update the initial feature vector according to the attention weight to obtain the attention feature vector of the device.

[0073] For example, the initial feature vector of the device can be transformed into a query, key, and value matrix through linear transformation, and the attention score, i.e. the attention weight, can be calculated. The attention weight and the value matrix are then weighted and summed to generate the attention feature vector of the device.

[0074] Through steps S4311 to S4312, key features are dynamically focused using attention weights, thus resolving the issue of unreasonable feature weight allocation. Joint optimization of motion feature initialization and attention weight updates facilitates efficient extraction of fatigue features from components. Compared to traditional methods, this approach improves feature quality and reduces fatigue life prediction errors.

[0075] Optionally, please refer to Figure 2 S440, based on the fatigue feature vector and the transformed feature vector, predicts the crack propagation path, including: S441, define the vertices of the mesh after the pair is meshed as nodes.

[0076] For example, the mating parts (needle valve and needle valve body) can be meshed using finite element analysis (FEA) or geometric modeling tools (such as ANSYS, ABAQUS), discretizing the continuous structure into a finite number of mesh elements, and defining the mesh vertices as nodes.

[0077] S442: For each node, aggregate the fatigue feature vectors of neighboring nodes and combine them with the transformed feature vectors to predict the crack propagation probability of each node.

[0078] For example, for each node, the fatigue feature vectors of neighboring nodes can be aggregated using a weighted average or attention mechanism, and then concatenated with the transformed feature vectors before using a multilayer perceptron (MLP) to predict the crack propagation probability.

[0079] S443, the crack propagation path is obtained based on the probability of each crack propagation.

[0080] For example, crack propagation paths can be generated by combining crack propagation probabilities with a probabilistic graphical model or a shortest path algorithm (such as Dijkstra's algorithm). For instance, for node i, its candidate propagation direction is the direction of its neighboring nodes. The crack propagation probability is converted into path cost. The stress concentration region is determined based on the stress distribution in the fused features, and the initial crack is determined based on the stress concentration region. Starting from the node of the initial crack, the lowest cost path to the node of the critical crack is searched to obtain the crack propagation path.

[0081] Through steps S441 to S443, the continuous problem is discretized using meshing, which improves computational efficiency. Probabilistic weighting and path search resolve the contradiction between the randomness of crack propagation paths and global optimization. By defining meshed nodes, aggregating probabilistic features, and searching for the optimal path, high-precision prediction of crack propagation in paired components is achieved. This helps shorten the crack propagation path prediction time, reduce prediction errors, and provides a reliable tool for assessing the fatigue life of paired components.

[0082] In one possible implementation, please refer to Figure 2S500 calculates the fatigue index based on geometric and kinematic features, and generates a failure competition map based on the fatigue index, including: S510: Construct a correlation network between geometric and motion features. This correlation network is used to analyze the correlation and influence between the components.

[0083] For example, a graph neural network (GNN) or spatial attention mechanism can be used to construct an association network between geometric features and motion features based on the topological relationships of the components in the fuel injector. Nodes in the association network are components, with geometric features of the components as node attributes and motion features as edge weights.

[0084] S520 calculates the fatigue index of each component based on the interconnected network.

[0085] For example, for each node in the associated network, the geometric and motion features of the neighboring nodes can be aggregated to generate a fatigue feature vector, which can then be mapped to a fatigue index using a multilayer perceptron (MLP).

[0086] S530 calculates the probability of each failure mode based on the fatigue index and predicts the corresponding failure time window.

[0087] For example, a mapping function from the fatigue index to the probability of occurrence of each failure mode can be established based on historical failure data, or the probability of failure mode can be calculated using probabilistic network estimation (PNET) based on the fatigue index. A Weibull distribution is fitted to the time data of each failure mode in the historical failure data to obtain the failure time window.

[0088] Through steps S510 to S530, the dynamic interaction between geometric deformation and moving loads is captured by the correlation network, which helps reduce crack propagation prediction errors. Multi-feature fusion improves the resolution of fatigue indices under complex working conditions. By combining the correlation network, multi-feature fusion, and probabilistic simulation, accurate prediction of fatigue failure of components is achieved, which improves the consistency of crack propagation paths, shortens computation time, and provides a reliable tool for component life assessment.

[0089] In one possible implementation, please refer to Figure 2 S100, acquires the geometric and kinematic features of the fuel injector, including: S110 uses simulation analysis to simulate the motion state of the fuel injector under different operating conditions and extracts the motion characteristics of each component.

[0090] For example, a 3D model can be built in multibody dynamics simulation software (such as ADAMS, RecurDyn, or SIMULIA) based on the CAD design data of the fuel injector (such as STEP or IGES format). Material parameters (such as density, elastic modulus, and Poisson's ratio) are assigned to each component. Based on predefined loads and constraints according to actual operating conditions, an explicit dynamics solver (such as ADAMS / Solver) is used to handle high-speed impacts (such as collisions during needle valve opening / closing) and transient responses, simulating fuel pressure fluctuations under different operating conditions (such as idling, full load, etc.). A simulation task is submitted to determine the motion state of the fuel injector within multiple cycles (such as 10 ms per cycle). Stress cloud diagrams are plotted in simulation post-processing software (such as HyperView or ADAMS / PostProcessor), outputting data such as stress distribution, deformation (such as needle valve displacement and valve seat elastic deformation), and vibration frequencies (analyzed by Fast Fourier Transform (FFT) to identify the dominant frequency), extracting the motion characteristics of each component.

[0091] Through step S110 above, the influence of different structural parameters on motion stability can be analyzed through simulation, which is beneficial for optimizing structural parameters. The motion characteristics of the fuel injector extracted by simulation provide high-precision input for subsequent fatigue analysis, which helps improve the reliability of component life prediction.

[0092] In one possible implementation, please refer to Figure 2 S600, the fatigue life of the component is obtained based on the crack propagation path and failure competition diagram, including: S610 converts the crack propagation path into a time series to obtain the crack propagation time.

[0093] For example, the crack propagation path can be discretized into a series of nodes, each node corresponding to the position of the crack tip and the propagation step size. According to the crack propagation rate model (such as the Paris formula), the crack length is converted into the number of cycles through numerical integration (such as the trapezoidal method). Then, combined with the loading frequency f (such as f=10Hz), the number of cycles is converted into time. The time increment corresponding to each propagation step size is calculated to obtain the crack propagation time.

[0094] S620, based on the failure competition diagram, corrects the crack propagation time to obtain the fatigue life.

[0095] For example, the crack propagation time can be corrected based on the Pareto diagram. The probability P of fatigue fracture is obtained from the Pareto diagram. fatigue Other failure mode probabilities P other and the time window T for each failure mode otherThe competing failure correction factor α (used to reflect the reduction in fatigue life caused by other failure modes) is calculated based on the probability of fatigue fracture, the probabilities of other failure modes, and the time windows of each failure mode. For example, α = 1− other I(N fatigue ∈T other ), where I is the indicator function, N fatigue ∈T other The value is 1 if the condition is met, and 0 otherwise. The fatigue life of the component is calculated based on the crack propagation time and the competing failure correction factor. Figure 4 As shown.

[0096] Through steps S610 to S620, a full-process simulation analysis from crack propagation path to fatigue life is achieved, which helps to solve problems such as high experimental costs, low data resolution, and insufficient operating condition coverage in existing technologies. By integrating multiple modes of influence through a failure competition graph, the reliability of predictions is improved. The failure competition graph can identify the dominant failure mode, which helps guide the structural optimization of fuel injectors.

[0097] In one possible implementation, please refer to Figure 2 The methods also include: S001, determine the structural parameters of the fuel injector within a preset range based on fatigue life.

[0098] It is understandable that structural parameters may include the needle valve cone angle, nozzle diameter, and needle valve body pressure chamber volume.

[0099] For example, with the goal of maximizing fatigue life and the constraint condition being a preset range of structural parameters, multiple sets of parameter combinations can be generated using a genetic algorithm (GA) or particle swarm optimization (PSO) to determine the geometric features of the fuel injector. Based on the changes in the geometric features of the fuel injector, the simulation can continue according to the above steps to obtain the structural parameters corresponding to the maximum fatigue life of the fuel injector assembly.

[0100] Through step S001 above, the structural parameters of the fuel injector are obtained through simulation, which helps reduce material consumption and optimize the fatigue life of the fuel injector. The algorithm avoids local optima through global search, which helps improve optimization efficiency.

[0101] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0102] Corresponding to the fatigue life simulation analysis method for engine injector assembly structure described in the above embodiments, this application also provides an engine injector assembly structure fatigue life simulation analysis device. The various modules of this device can realize the various steps of the engine injector assembly structure fatigue life simulation analysis method. Figure 5 The diagram shows a structural block diagram of the fatigue life simulation analysis device for engine injector assembly provided in the embodiments of this application. For ease of explanation, only the parts related to the embodiments of this application are shown.

[0103] Reference Figure 5 The device includes: The acquisition module is used to acquire the geometric and kinematic features of the fuel injector; wherein, the geometric features include size, shape and surface roughness, and the kinematic features are used to reflect the deformation and vibration of the fuel injector under different operating conditions; The position information module is used to establish a spatial correlation matrix of each component in the fuel injector based on the geometric features to obtain position information; wherein, the components include a coupler, a spring, an O-ring, and a filter screen, and the coupler includes a needle valve and a needle valve body; A feature fusion module is used to fuse the motion features and the position information to obtain fused features; A crack propagation path module is used to predict the crack propagation path of the pair based on the fusion features. The failure competition graph module is used to calculate the fatigue index based on the geometric features and the motion features, and generate a failure competition graph based on the fatigue index; wherein, the failure competition graph includes the probabilities of multiple failure modes and the corresponding failure time windows; The fatigue life module is used to obtain the fatigue life of the component based on the crack propagation path and the failure competition diagram.

[0104] It should be noted that the information interaction and execution process between the above modules are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0105] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above device can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0106] This application also provides an electronic device. Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 6 As shown, the electronic device 6 of this embodiment includes: at least one processor 60 ( Figure 6 Only one is shown in the image), at least one memory 61 ( Figure 6 (Only one is shown in the image) and a computer program 62 stored in the at least one memory 61 and executable on the at least one processor 60. When the processor 60 executes the computer program 62, it causes the electronic device 6 to implement the steps in any of the above embodiments of the simulation analysis method for fatigue life of engine injector assembly structure, or to implement the functions of each module / unit in the above embodiments of the device.

[0107] For example, the computer program 62 may be divided into one or more modules / units, which are stored in the memory 61 and executed by the processor 60 to complete this application. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program 62 in the electronic device 6.

[0108] The electronic device 6 can be a computing device such as an industrial computer, programmable logic controller, desktop computer, laptop, handheld computer, and cloud server. This electronic device may include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art will understand that... Figure 6This is merely an example of electronic device 6 and does not constitute a limitation on electronic device 6. It may include more or fewer components than shown, or combine certain components, or different components, such as input / output devices, network access devices, buses, etc.

[0109] The processor 60 can be a Central Processing Unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0110] In some embodiments, the memory 61 may be an internal storage unit of the electronic device 6, such as a hard disk or memory of the electronic device 6. In other embodiments, the memory 61 may be an external storage device of the electronic device 6, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the electronic device 6. Furthermore, the memory 61 may include both internal and external storage units of the electronic device 6. The memory 61 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory 61 can also be used to temporarily store data that has been output or will be output.

[0111] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0112] This application provides a computer program product that, when run on an electronic device, causes the electronic device to perform the steps in any of the above method embodiments.

[0113] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to an electronic device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0114] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0115] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0116] In the embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the electronic device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0117] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0118] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for simulation analysis of fatigue life of engine fuel injector assembly structure, characterized in that, include: The geometric and kinematic features of the fuel injector are obtained; wherein the geometric features include size, shape and surface roughness, and the kinematic features are used to reflect the deformation and vibration of the fuel injector under different operating conditions. Based on the geometric features, a spatial correlation matrix of each component in the fuel injector is established to obtain position information; wherein, the components include a coupler, a spring, an O-ring, and a filter screen, and the coupler includes a needle valve and a needle valve body; The motion features and the location information are fused to obtain fused features; The crack propagation path of the pair is predicted based on the fusion features; A fatigue index is calculated based on the geometric features and the motion features, and a failure competition map is generated based on the fatigue index; wherein, the failure competition map includes the probabilities of multiple failure modes and the corresponding failure time windows; The fatigue life of the component is obtained based on the crack propagation path and the failure competition diagram.

2. The fatigue life simulation analysis method for engine fuel injector assembly structure as described in claim 1, characterized in that, The step of predicting the crack propagation path of the matte component based on the fused features includes: A fatigue feature vector is obtained based on the fusion features; wherein, the fatigue feature vector is used to quantify the fatigue sensitivity of the component under cyclic loading; The attention weights for the fatigue life of the component are obtained based on the fatigue feature vector; wherein the attention weights are used to reflect the degree of contribution of each component in the fatigue feature vector to the prediction of the crack propagation path. The transformation feature vector is determined based on the attention weight; wherein the transformation feature vector is used to reflect the degree of fatigue influence between the needle valve and the needle valve body in the coupling. The crack propagation path is obtained by predicting based on the fatigue feature vector and the transformed feature vector.

3. The fatigue life simulation analysis method for engine fuel injector assembly structure as described in claim 2, characterized in that, The step of obtaining the fatigue feature vector based on the fused features includes: Based on the fusion features of the device, a first correlation between the needle valve and the needle valve body in the device is calculated to obtain a first correlation vector; wherein, the first correlation vector is used to reflect the linear correlation between the needle valve and the needle valve body in the fusion feature space of the device. Based on the first correlation vector and the fusion feature, the second correlation between the needle valve and the needle valve body in the pair is calculated to obtain the second correlation vector; wherein, the second correlation vector is used to reflect the nonlinear correlation between the needle valve and the needle valve body in the fusion feature space of the pair; The fatigue feature vector of the device is obtained by fusing the first correlation vector with the second correlation vector.

4. The fatigue life simulation analysis method for engine fuel injector assembly structure as described in claim 2, characterized in that, The process of determining the transformed feature vector based on the attention weights includes: The attention feature vectors of the needle valve and the needle valve body in the device are determined based on the attention weights. The transformation feature vectors of the needle valve and the needle valve body in the device are obtained based on the attention feature vector.

5. The fatigue life simulation analysis method for engine fuel injector assembly structure as described in claim 4, characterized in that, Determining the attention feature vectors of the needle valve and needle valve body in the coupling based on the attention weights includes: The initial feature vector of the device is obtained based on the motion characteristics; The initial feature vector is updated according to the attention weights to obtain the attention feature vector of the device.

6. The fatigue life simulation analysis method for engine fuel injector assembly structure as described in claim 2, characterized in that, The step of predicting the crack propagation path based on the fatigue feature vector and the transformed feature vector includes: The vertices of the meshed component are defined as nodes; For each node, the fatigue feature vectors of the neighboring nodes are aggregated, and the crack propagation probability of each node is predicted by combining the transformed feature vectors. The crack propagation path is obtained based on the crack propagation probability.

7. The fatigue life simulation analysis method for engine fuel injector assembly structure as described in claim 1, characterized in that, The step of calculating the fatigue index based on the geometric features and the motion features, and generating a failure competition map based on the fatigue index, includes: Construct a correlation network between the geometric features and the motion features; wherein the correlation network is used to analyze the correlation and degree of influence between the components. The fatigue index of each component is calculated based on the associated network; The probability of each failure mode is calculated based on the fatigue index, and the corresponding failure time window is predicted.

8. The fatigue life simulation analysis method for engine fuel injector assembly as described in claim 1, characterized in that, The acquisition of the geometric and kinematic features of the fuel injector includes: Through simulation analysis, the motion state of the fuel injector under different operating conditions is simulated, and the motion characteristics of each component are extracted.

9. The fatigue life simulation analysis method for engine fuel injector assembly structure as described in claim 1, characterized in that, The step of obtaining the fatigue life of the mating component based on the crack propagation path and the failure competition diagram includes: The crack propagation path is converted into a time series to obtain the crack propagation time; The fatigue life is obtained by correcting the crack propagation time based on the failure competition diagram.

10. The fatigue life simulation analysis method for engine fuel injector assembly structure as described in claim 1, characterized in that, The method further includes: The structural parameters of the fuel injector are determined within a preset range based on the fatigue life.

Images