Optimization method based on influence of shear stress amplitude on cyclic loading characteristic of bonding structure

Abstract

The invention discloses an optimization method based on the influence of shear stress amplitude on the cyclic loading characteristic of a bonding structure, relates to the technical field of bonding technology and connection engineering, and aims at maximizing an effective fatigue performance index (EFPI) by constructing a comprehensive evaluation model comprising a geometric stress release index (GSRI), a process damage potential index (PDPI) and a damage invisibility index (DI). And in the initial stage of design, geometric income, manufacturing cost and microdefects and leak detection risks introduced by the process are quantitatively weighed. According to the method, a high-cost and high-risk complex geometric scheme can be actively avoided, an optimal solution considering performance and economical efficiency is automatically screened through a decision threshold value, and low-risk alternative strategies such as functional gradient adhesives are activated and adopted when no optimal solution exists. Finally, the manufacturing cost is remarkably reduced, the paradox of actual performance degradation caused by process defects is fundamentally avoided, and the reliability, safety and engineering application value of optimization design are comprehensively improved.

Description

Technical Field

[0001] This invention relates to the fields of adhesive technology and connection engineering technology, specifically to an optimization method based on the influence of shear stress amplitude on the cyclic loading characteristics of bonded structures. Background Technology

[0002] Adhesive bonding technology, due to its advantages of lightweight design and uniform stress distribution, is increasingly widely used in high-end manufacturing fields such as aerospace and automotive. However, fatigue failure under cyclic loading remains a key challenge. Early designs focused on static strength, but with the development of numerical techniques such as finite element analysis, research has delved into the dynamic characteristics. It has been found that the concentration of shear stress amplitude at the lap joint is the source of fatigue cracks. Therefore, methods to regulate shear stress distribution and extend the service life of structures by optimizing joint geometry (such as end thinning) and material properties have emerged and have become the core guiding principle for the reliability design of modern adhesive structures.

[0003] In practical applications, this optimization method has the following two related technical drawbacks: To achieve the theoretically optimal shear stress distribution, complex geometric modifications to the bonded components are often required, such as precise end thinning or non-linear curve overlapping. This necessitates advanced but expensive manufacturing processes such as multi-axis CNC machining and precision mold forming, leading to extended production cycles, increased equipment investment, and higher scrap rates. Consequently, manufacturing costs are significantly increased, limiting its large-scale application in cost-sensitive products.

[0004] Furthermore, process-introduced micro-defects become new sources of fatigue, which is closely related to the aforementioned drawbacks but is more insidious. During machining processes such as thinning of bonded parts (especially composite materials), the cutting process itself may introduce new micro-defects into the optimized area, such as matrix microcracks, fiber fractures, or interface delamination. These "process-introduced defects" may be more dangerous than the original geometric stress concentration points, becoming preferred sites for fatigue crack initiation. Ultimately, a paradox may emerge: the manufacturing process employed to eliminate geometric stress concentrations may inadvertently create more fatal material-level defects, resulting in actual fatigue performance of the structure far below the theoretical expectations of the optimized design. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an optimization method based on the influence of shear stress amplitude on the cyclic loading characteristics of bonded structures, thus resolving the technical deficiencies mentioned in the background section.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an optimization method based on the influence of shear stress amplitude on the cyclic loading characteristics of bonded structures, comprising the following steps: S1. Perform finite element modeling on the initial bonded structure and apply cyclic loads to analyze and obtain the baseline distribution of the shear stress amplitude of the adhesive layer, and extract the baseline maximum shear stress amplitude. S2. For the stress concentration region where the reference maximum shear stress amplitude is located, set a geometric optimization scheme and perform finite element analysis again to obtain the optimized maximum shear stress amplitude. Based on the difference between the reference maximum shear stress amplitude and the maximum shear stress amplitude, calculate the geometric stress release index GSRI, which characterizes the theoretical benefit of the scheme. S3. Evaluate the manufacturing process required to implement the geometry optimization scheme, and comprehensively consider the process complexity, material sensitivity and geometric severity to calculate the process damage potential index (PDPI) that quantifies the risk of microscopic damage introduced by the process. S4. Based on the selected manufacturing process, evaluate the effectiveness of the non-destructive testing scheme for detecting potential micro-damage, and calculate the damage invisibility index (DII) to characterize the risk of missed defects. S5, along with the associated geometric stress release index GSRI, process damage potential index PDPI, and damage invisibility index DII, constructs an effective fatigue performance index EFPI for comprehensively evaluating the actual fatigue performance of the structure. Based on the evaluation results of this index, the final decision is made on the geometric optimization scheme and manufacturing process.

[0007] Preferably, the Geometric Stress Relief Index (GSRI) mentioned in S2 aims to evaluate the benefits of geometric optimization purely from a mechanical design perspective, and its specific calculation logic features are as follows: Obtain the maximum shear stress amplitude of the unoptimized initial bonded structure as the evaluation benchmark; obtain the optimized maximum shear stress amplitude of the bonded structure under the same load conditions after adopting the geometric optimization scheme; Calculate the absolute difference between the initial maximum shear stress amplitude and the optimized maximum shear stress amplitude. This difference represents the absolute amount by which the stress peak is reduced. Compare this absolute difference with the initial maximum shear stress amplitude to obtain a normalized ratio. This ratio is the Geometric Stress Relief Index (GSRI), and its value directly reflects the relative degree to which the theoretical stress concentration is alleviated.

[0008] Preferably, the Process Damage Potential Index (PDPI) mentioned in S3 aims to quantify the possibility of the manufacturing process introducing microscopic defects into the structural matrix. The calculation logic of the PDPI is as follows: Based on the inherent properties of the selected manufacturing process, such as the invasiveness and control precision of the processing principle, a process complexity coefficient value is determined; based on the physical properties of the materials being processed, especially their tolerance and sensitivity to damage from machining, a material damage sensitivity coefficient value is determined; based on the degree of change in the geometric shape in the optimized scheme, such as the slope of the thinned region or the curvature of the transition curve, a geometric gradient coefficient value is determined. The product of the determined process complexity coefficient, material damage sensitivity coefficient, and geometric gradient coefficient is the process damage potential index (PDPI).

[0009] Preferably, the Damage Invisibility Index (DII) described in S4 aims to quantify the risk of failing to detect critical microscopic defects due to limitations in detection methods. The DII quantifies the risk of failing to detect critical process defects by evaluating the complement of the effectiveness of the selected non-destructive testing scheme. The specific calculation logic is as follows: Based on the estimated types and sizes of micro-defects, the detection capabilities of the selected non-destructive testing scheme are comprehensively evaluated, including its detection coverage and accuracy, so as to determine a non-destructive testing effectiveness coefficient value between 0 and 1, where 1 represents theoretically perfect detection capability; Subtract the determined nondestructive testing effectiveness coefficient from the value 1, which represents complete testing capability; the resulting difference is the damage invisibility index (DII), and the magnitude of the damage invisibility index (DII) is directly proportional to the risk of missing defects.

[0010] Preferably, the Effective Fatigue Performance Index (EFPI) mentioned in S5 aims to predict the overall performance of a structure under actual working conditions by quantifying the trade-off between benefits and dual risks, namely the risk of generation and the risk of missed detection. The calculation logic of the Effective Fatigue Performance Index (EFPI) is as follows: The first step is to calculate the weighted geometric return by multiplying the geometric stress relief index (GSRI) by the preset geometric return weighting coefficient. The second step is to calculate the weighted risk of loss by multiplying the process loss potential index PDPI by the preset loss potential weight coefficient. This is the first risk penalty item. The third step is to calculate the weighted risk of missed detection: First, multiply the process damage potential index PDPI with the damage invisibility index DII to obtain the coupled risk value that represents "invisible damage". Then, multiply the coupled risk value with the preset missed detection risk weight coefficient. This is the second risk penalty term. A comprehensive calculation is performed, subtracting the weighted risk of loss calculated in the second step and the weighted risk of missed detection calculated in the third step from the weighted geometric benefit calculated in the first step; the final calculation result is the Effective Fatigue Performance Index (EFPI), which is used for the final decision-making.

[0011] Preferably, the final decision on the geometry optimization scheme and manufacturing process in S5 is specifically as follows: the calculated effective fatigue performance index EFPI is compared with a preset decision threshold, wherein the decision threshold includes a first decision threshold Ac and a second decision threshold Ar, and the first decision threshold Ac > the second decision threshold Ar. If the effective fatigue performance index EFPI > the first decision threshold Ac, the scheme is determined to be a high-yield, low-risk scheme, and the geometric design and supporting manufacturing and testing processes are approved for adoption. If the second decision threshold Ar < effective fatigue performance index EFPI ≤ first decision threshold Ac, it is determined to be a critical solution with comparable benefits and risks, triggering a warning and requiring an increase in the frequency of non-destructive testing or a re-evaluation after small-batch process verification. If the effective fatigue performance index EFPI is less than or equal to the second decision threshold Ar, the design is deemed high-risk, the geometric design is forcibly rejected, and the process returns to step S2 for redesign.

[0012] Preferably, during the redesign in step S2, the ultimate optimization objective is to maximize the effective fatigue performance index EFPI, and the geometric gradient coefficient Kg and process complexity coefficient Kp are used as design variables. Iterative optimization is performed within the preset cost and process capability boundaries to find a balanced solution that is not theoretically optimal but has the best overall performance.

[0013] Preferably, when the maximum effective fatigue performance index EFPI after iterative optimization still cannot exceed the second decision threshold Ar, the system will automatically lock and abandon the complex geometric optimization path and activate the alternative optimization strategy library.

[0014] Preferably, the alternative optimization strategy library includes: Strategy 1: Keep the original or simplified geometry unchanged, and instead optimize the properties of the adhesive material, such as using functionally graded adhesives or flexible adhesives at the ends; Strategy 2: Based on the original geometric configuration, adopt low-invasive and low-process-risk structural reinforcement measures such as increasing the overlap length or adding external reinforcing patches.

[0015] Preferably, the method further includes constructing an associated database to store the geometric schemes, process parameters, non-destructive testing schemes, and their corresponding calculated values ​​of the Geometric Stress Relief Index (GSRI), Damage Invisibility Index (DII), and Process Damage Potential Index (PDPI) for historical projects, as well as the actual structural life verified by fatigue experiments. A machine learning model is then used to train this database to continuously correct the calibration values ​​of the process complexity coefficient Kp and the material damage sensitivity coefficient Km, and to optimize the weighting coefficients, including the allocation strategy for the geometric benefit weighting coefficient, damage potential weighting coefficient Wp, and missed detection risk weighting coefficient Wd, thereby improving the accuracy and efficiency of subsequent optimization decisions.

[0016] This invention provides an optimization method based on the influence of shear stress amplitude on the cyclic loading characteristics of bonded structures. It has the following beneficial effects: (1) The optimization method based on the influence of shear stress amplitude on the cyclic loading characteristics of bonded structures systematically solves the technical problem of uncontrolled manufacturing costs caused by pursuing theoretically optimal geometry by constructing a comprehensive evaluation system including geometric stress release index GSRI, process damage potential index PDPI and damage invisibility index DII, and taking the maximization of effective fatigue performance index EFPI as the optimization goal. The method introduces quantitative evaluation of process complexity coefficient Kp and geometric gradient coefficient Kg in the early stage of design, so that the optimization process no longer blindly pursues the ultimate smoothness of stress distribution, but actively avoids complex geometric configurations that require high-cost manufacturing processes such as multi-axis CNC machining. By setting decision thresholds Ac and Ar, this method can automatically screen and reject design schemes that are not enough to offset their high manufacturing costs, guide the design to converge towards a more cost-effective direction, thereby significantly reducing the production cycle, equipment investment and scrap rate while ensuring that the structural performance meets the requirements, and greatly enhancing the applicability and economic value of the optimized design in cost-sensitive products.

[0017] (2) This optimization method based on the influence of shear stress amplitude on the cyclic loading characteristics of bonded structures, by innovatively establishing the process damage potential index PDPI and the damage invisibility index DII, profoundly solves the hidden technical pain point in optimization design that "process-introduced defects become more dangerous fatigue sources"; this method not only quantifies the risk of microcracks or fiber damage introduced by complex processing operations, but also proactively incorporates the non-destructive testing difficulty of these potential defects into the decision model. By calculating the effective fatigue performance index EFPI containing coupled risk terms, it accurately identifies and punishes those "high-risk and difficult-to-detect" design schemes; when iterative optimization cannot find a satisfactory solution, the method can automatically abandon high-risk geometric modification paths and instead activate alternative strategies with low intrusion and low process risk, such as using functionally graded adhesives or increasing the overlap length. This fundamentally avoids the paradox of creating more fatal material defects in order to eliminate theoretical stress concentration, ensuring that the actual fatigue performance of the optimized structure meets or even exceeds theoretical expectations, and significantly improves the final reliability and safety of the product. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the steps of the optimization method of the present invention based on the influence of shear stress amplitude on the cyclic loading characteristics of bonded structures; Detailed Implementation The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1 Please see Figure 1 This invention provides an optimization method based on the influence of shear stress amplitude on the cyclic loading characteristics of bonded structures. S1, finite element modeling is performed on the initial bonded structure and cyclic load is applied. The baseline distribution of shear stress amplitude of the adhesive layer is analyzed and obtained, and the baseline maximum shear stress amplitude is extracted. S2. For the stress concentration region where the reference maximum shear stress amplitude is located, set a geometric optimization scheme and perform finite element analysis again to obtain the optimized maximum shear stress amplitude. Based on the difference between the reference maximum shear stress amplitude and the maximum shear stress amplitude, calculate the geometric stress release index GSRI, which characterizes the theoretical benefit of the scheme. S3. Evaluate the manufacturing process required to implement the geometry optimization scheme, and comprehensively consider the process complexity, material sensitivity and geometric severity to calculate the process damage potential index (PDPI) that quantifies the risk of microscopic damage introduced by the process. S4. Based on the selected manufacturing process, evaluate the effectiveness of the non-destructive testing scheme for detecting potential micro-damage, and calculate the damage invisibility index (DII) to characterize the risk of missed defects. S5, along with the associated geometric stress release index GSRI, process damage potential index PDPI, and damage invisibility index DII, constructs an effective fatigue performance index EFPI for comprehensively evaluating the actual fatigue performance of the structure. Based on the evaluation results of this index, the final decision is made on the geometric optimization scheme and manufacturing process.

[0020] In this embodiment, step S1 extracts the baseline maximum shear stress amplitude, which can accurately locate the weak points in the structure's fatigue and provide a quantitative basis for optimization. Then, step S2 calculates the Geometric Stress Relief Index (GSRI), which allows for a standardized comparison of the theoretical benefits of different geometric schemes in reducing stress concentration. Next, step S3 calculates the Process Damage Potential Index (PDPI), which innovatively quantifies the risk of hidden damage introduced by the manufacturing process, making up for the shortcomings of traditional optimization that are divorced from engineering reality. Then, step S4 calculates the Damage Invisibility Index (DII), which can proactively assess and avoid those "high-risk and difficult-to-detect" fatal schemes, improving product safety. Finally, step S5 constructs a comprehensive Effective Fatigue Performance Index (EFPI) for decision-making, which unifies the benefits represented by the Geometric Stress Relief Index (GSRI) with the dual risks represented by the Process Damage Potential Index (PDPI) and the Damage Invisibility Index (DII), achieving a scientific, efficient, and comprehensive scheme evaluation and selection.

[0021] Example 2 The Geometric Stress Relief Index (GSRI) described in S2 aims to evaluate the benefits of geometric optimization purely from a mechanical design perspective. Its specific calculation logic is as follows: Obtain the maximum shear stress amplitude of the unoptimized initial bonded structure as the evaluation benchmark; obtain the optimized maximum shear stress amplitude of the bonded structure under the same load conditions after adopting the geometric optimization scheme; The absolute difference between the initial maximum shear stress amplitude and the optimized maximum shear stress amplitude is calculated. This difference represents the absolute amount by which the stress peak is reduced. This absolute difference is compared with the initial maximum shear stress amplitude to obtain a normalized ratio, which is the geometric stress relief index (GSRI). Its value directly reflects the relative degree to which the theoretical stress concentration is alleviated. The GSRI value range is [0, 1]. The larger the value, the more significant the effect of the geometric optimization scheme in reducing the theoretical stress concentration.

[0022] The Process Damage Potential Index (PDPI) described in S3 aims to quantify the likelihood of the manufacturing process introducing microscopic defects into the structural matrix. The calculation logic for the PDPI is as follows: Based on the inherent properties of the selected manufacturing process, such as the invasiveness and control precision of the processing principle, a process complexity coefficient value is determined; based on the physical properties of the materials being processed, especially their tolerance and sensitivity to damage from machining, a material damage sensitivity coefficient value is determined; based on the degree of change in the geometric shape in the optimized scheme, such as the slope of the thinned region or the curvature of the transition curve, a geometric gradient coefficient value is determined. Multiplying the determined process complexity coefficient, material damage sensitivity coefficient, and geometric gradient coefficient values ​​together yields the process damage potential index (PDPI). This calculation logic ensures that a significant increase in risk in any dimension will lead to a multiple-fold increase in the final comprehensive potential index.

[0023] The process complexity coefficient is calibrated based on the processing method, including the invasiveness and control precision of three-axis or five-axis cutting, precision casting, and additive manufacturing; the material damage sensitivity coefficient is calibrated based on the type of the adhered material, including isotropic metals and orthotropic composite materials and their micro-damage tolerance; the geometric gradient coefficient is calibrated based on the rate of curvature change or thinning slope of the optimized region, characterizing the severity of the processing.

[0024] The Damage Invisibility Index (DII) described in S4 aims to quantify the risk of failing to detect critical microscopic defects due to limitations in detection methods. The DII quantifies the risk of failing to detect critical process defects by evaluating the complement of the effectiveness of the selected non-destructive testing (NDT) scheme. The specific calculation logic is as follows: Based on the estimated types and sizes of micro-defects, the detection capabilities of the selected non-destructive testing scheme are comprehensively evaluated, including its detection coverage and accuracy, so as to determine a non-destructive testing effectiveness coefficient value between 0 and 1, where 1 represents theoretically perfect detection capability; Subtract the determined nondestructive testing effectiveness coefficient from the value 1, which represents complete testing capability; the resulting difference is the damage invisibility index (DII), and the magnitude of the damage invisibility index (DII) is directly proportional to the risk of missing defects.

[0025] The Effective Fatigue Performance Index (EFPI) described in S5 aims to predict the overall performance of a structure under actual working conditions by quantifying the trade-off between benefits and dual risks, namely the risk of generation and the risk of missed detection. The calculation logic of the Effective Fatigue Performance Index (EFPI) is as follows: The first step is to calculate the weighted geometric return by multiplying the geometric stress relief index (GSRI) by the preset geometric return weighting coefficient. The second step is to calculate the weighted risk of loss by multiplying the process loss potential index PDPI by the preset loss potential weight coefficient. This is the first risk penalty item. The third step is to calculate the weighted risk of missed detection: First, multiply the process damage potential index PDPI with the damage invisibility index DII to obtain the coupled risk value that represents "invisible damage". Then, multiply the coupled risk value with the preset missed detection risk weight coefficient. This is the second risk penalty term. A comprehensive calculation is performed, subtracting the weighted risk of loss calculated in the second step and the weighted risk of missed detection calculated in the third step from the weighted geometric benefit calculated in the first step; the final calculation result is the Effective Fatigue Performance Index (EFPI), which is used for the final decision-making.

[0026] The final decision on the geometry optimization scheme and manufacturing process in S5 is as follows: the calculated effective fatigue performance index EFPI is compared with the preset decision threshold, the decision threshold includes a first decision threshold Ac and a second decision threshold Ar, and the first decision threshold Ac > the second decision threshold Ar. If the effective fatigue performance index EFPI > the first decision threshold Ac, the scheme is determined to be a high-yield, low-risk scheme, and the geometric design and supporting manufacturing and testing processes are approved for adoption. If the second decision threshold Ar < effective fatigue performance index EFPI ≤ first decision threshold Ac, it is determined to be a critical solution with comparable benefits and risks, triggering a warning and requiring an increase in the frequency of non-destructive testing or a re-evaluation after small-batch process verification. If the effective fatigue performance index EFPI is less than or equal to the second decision threshold Ar, the design is deemed high-risk, the geometric design is forcibly rejected, and the process returns to step S2 for redesign.

[0027] During the redesign in step S2, the ultimate optimization objective is to maximize the effective fatigue performance index EFPI. The geometric gradient coefficient Kg and the process complexity coefficient Kp are used as design variables. The optimization is carried out iteratively within the preset cost and process capability boundaries to find a balance solution that is not theoretically optimal but has the best overall performance.

[0028] When the maximum effective fatigue performance index EFPI after iterative optimization still cannot exceed the second decision threshold Ar, the system will automatically lock and abandon the complex geometric optimization path and activate the alternative optimization strategy library.

[0029] The alternative optimization strategy library includes: Strategy 1: Keep the original or simplified geometry unchanged, and instead optimize the properties of the adhesive material, such as using functionally graded adhesives or flexible adhesives at the ends; Strategy 2: Based on the original geometric configuration, adopt low-invasive and low-process-risk structural reinforcement measures such as increasing the overlap length or adding external reinforcing patches.

[0030] In this embodiment, step S5 further defines the calculation and decision-making method for the Effective Fatigue Performance Index (EFPI). This method first uses the Analytic Hierarchy Process (AHP) combined with historical engineering data to assign weights to geometric gains, process-induced damage potential, and damage omission risk, obtaining the geometric gain weight coefficient Wg, damage potential weight coefficient Wp, and omission risk weight coefficient Wd, respectively. Subsequently, a comprehensive calculation is performed based on the following formula:

[0031] The formula includes a weighted geometric benefit term and two risk penalty terms: the first is the weighted damage risk, representing the direct risk of the process; the second is the coupled risk, representing "invisible damage," obtained by multiplying the process damage potential index (PDPI) and the damage invisibility index (DII) and then weighting them. This calculation logic, by double-penalizing the process damage risk, particularly by non-linearly amplifying the negative impact of fatal defects with both high risk and high failure rate, allows the EFPI to more realistically predict the overall performance of the structure under actual operating conditions.

[0032] After obtaining the calculation result of the Effective Fatigue Performance Index (EFPI), this invention further defines an automated closed-loop decision-making process based on preset thresholds. This process compares the calculated EFPI value with preset first decision threshold Ac and second decision threshold Ar, where Ac is greater than Ar. If the EFPI value is higher than the first decision threshold Ac, the scheme is determined to be high-return and low-risk, and the system automatically approves and adopts the geometric design and supporting process. If the EFPI value is between the first and second decision thresholds, it is determined to be a critical scheme, and the system automatically triggers a warning and requires an increase in the frequency of non-destructive testing or process verification followed by re-evaluation. If the EFPI value is lower than the second decision threshold Ar, it is determined to be a high-risk scheme, and the system automatically executes a forced rejection instruction and returns the geometric design to the previous optimization step for redesign. This three-level adjudication mechanism transforms the quantitative evaluation results into clear engineering instructions, achieving objective, efficient, and automated screening and control of design schemes. Preferably, the method further includes constructing an associated database to store the geometric schemes, process parameters, non-destructive testing schemes, and their corresponding calculated values ​​of the Geometric Stress Relief Index (GSRI), Damage Invisibility Index (DII), and Process Damage Potential Index (PDPI) for historical projects, as well as the actual structural life verified by fatigue experiments. A machine learning model is then used to train this database to continuously correct the calibration values ​​of the process complexity coefficient Kp and the material damage sensitivity coefficient Km, and to optimize the weighting coefficients, including the allocation strategy for the geometric benefit weighting coefficient, damage potential weighting coefficient Wp, and missed detection risk weighting coefficient Wd, thereby improving the accuracy and efficiency of subsequent optimization decisions.

[0033] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An optimization method based on the influence of shear stress amplitude on the cyclic loading characteristics of bonded structures, characterized in that: Includes the following steps: S1. Perform finite element modeling on the initial bonded structure and apply cyclic loads to analyze and obtain the baseline distribution of the shear stress amplitude of the adhesive layer, and extract the baseline maximum shear stress amplitude. S2. For the stress concentration region where the reference maximum shear stress amplitude is located, set a geometric optimization scheme and perform finite element analysis again to obtain the optimized maximum shear stress amplitude. Based on the difference between the reference maximum shear stress amplitude and the maximum shear stress amplitude, calculate the geometric stress release index GSRI, which characterizes the theoretical benefit of the scheme. S3. Evaluate the manufacturing process required to implement the geometry optimization scheme, and comprehensively consider the process complexity, material sensitivity and geometric severity to calculate the process damage potential index (PDPI) that quantifies the risk of microscopic damage introduced by the process. S4. Based on the selected manufacturing process, evaluate the effectiveness of the non-destructive testing scheme for detecting potential micro-damage, and calculate the damage invisibility index (DII) to characterize the risk of missed defects. S5, along with the associated geometric stress release index GSRI, process damage potential index PDPI, and damage invisibility index DII, constructs an effective fatigue performance index EFPI for comprehensively evaluating the actual fatigue performance of the structure. Based on the evaluation results of this index, the final decision is made on the geometric optimization scheme and manufacturing process.

2. The optimization method based on the influence of shear stress amplitude on the cyclic loading characteristics of bonded structures according to claim 1, characterized in that: The Geometric Stress Relief Index (GSRI) described in S2 aims to evaluate the benefits of geometric optimization purely from a mechanical design perspective. Its specific calculation logic is as follows: Obtain the maximum shear stress amplitude of the unoptimized initial bonded structure as the evaluation benchmark; obtain the optimized maximum shear stress amplitude of the bonded structure under the same load conditions after adopting the geometric optimization scheme; Calculate the absolute difference between the initial maximum shear stress amplitude and the optimized maximum shear stress amplitude. This difference represents the absolute amount by which the stress peak is reduced. Compare this absolute difference with the initial maximum shear stress amplitude to obtain a normalized ratio. This ratio is the Geometric Stress Relief Index (GSRI), and its value directly reflects the relative degree to which the theoretical stress concentration is alleviated.

3. The optimization method based on the influence of shear stress amplitude on the cyclic loading characteristics of bonded structures according to claim 1, characterized in that: The Process Damage Potential Index (PDPI) described in S3 aims to quantify the likelihood of the manufacturing process introducing microscopic defects into the structural matrix. The calculation logic for the PDPI is as follows: Based on the inherent properties of the selected manufacturing process, such as the invasiveness and control precision of the processing principle, a process complexity coefficient value is determined; based on the physical properties of the materials being processed, especially their tolerance and sensitivity to damage from machining, a material damage sensitivity coefficient value is determined. A geometric gradient coefficient value is determined based on the degree of change in the geometric shape in the optimization scheme, such as the slope of the thinned region or the curvature of the transition curve. The product of the determined process complexity coefficient, material damage sensitivity coefficient, and geometric gradient coefficient is the process damage potential index (PDPI).

4. The optimization method based on the influence of shear stress amplitude on the cyclic loading characteristics of bonded structures according to claim 1, characterized in that: The Damage Invisibility Index (DII) described in S4 aims to quantify the risk of failing to detect critical microscopic defects due to limitations in detection methods. The DII quantifies the risk of failing to detect critical process defects by evaluating the complement of the effectiveness of the selected non-destructive testing (NDT) scheme. The specific calculation logic is as follows: Based on the estimated types and sizes of micro-defects, the detection capabilities of the selected non-destructive testing scheme are comprehensively evaluated, including its detection coverage and accuracy, so as to determine a non-destructive testing effectiveness coefficient value between 0 and 1, where 1 represents theoretically perfect detection capability; Subtract the determined nondestructive testing effectiveness coefficient from the value 1, which represents complete testing capability; the resulting difference is the damage invisibility index (DII), and the magnitude of the damage invisibility index (DII) is directly proportional to the risk of missing defects.

5. The optimization method based on the influence of shear stress amplitude on the cyclic loading characteristics of bonded structures according to claims 1, 3, and 4, characterized in that: The Effective Fatigue Performance Index (EFPI) described in S5 aims to predict the overall performance of a structure under actual working conditions by quantifying the trade-off between benefits and dual risks, namely the risk of generation and the risk of missed detection. The calculation logic of the Effective Fatigue Performance Index (EFPI) is as follows: The first step is to calculate the weighted geometric return by multiplying the geometric stress relief index (GSRI) by the preset geometric return weighting coefficient. The second step is to calculate the weighted risk of loss by multiplying the process loss potential index PDPI by the preset loss potential weight coefficient. This is the first risk penalty item. The third step is to calculate the weighted risk of missed detection: First, multiply the process damage potential index PDPI with the damage invisibility index DII to obtain the coupled risk value that represents "invisible damage". Then, multiply the coupled risk value with the preset missed detection risk weight coefficient. This is the second risk penalty term. A comprehensive calculation is performed, subtracting the weighted risk of loss calculated in the second step and the weighted risk of missed detection calculated in the third step from the weighted geometric benefit calculated in the first step; the final calculation result is the Effective Fatigue Performance Index (EFPI), which is used for the final decision-making.

6. The optimization method based on the influence of shear stress amplitude on the cyclic loading characteristics of bonded structures according to claim 5, characterized in that: The final decision on the geometry optimization scheme and manufacturing process in S5 is as follows: the calculated effective fatigue performance index EFPI is compared with the preset decision threshold, the decision threshold includes a first decision threshold Ac and a second decision threshold Ar, and the first decision threshold Ac > the second decision threshold Ar. If the effective fatigue performance index EFPI > the first decision threshold Ac, the scheme is determined to be a high-yield, low-risk scheme, and the geometric design and supporting manufacturing and testing processes are approved for adoption. If the second decision threshold Ar < effective fatigue performance index EFPI ≤ first decision threshold Ac, it is determined to be a critical solution with comparable benefits and risks, triggering a warning and requiring an increase in the frequency of non-destructive testing or a re-evaluation after small-batch process verification. If the effective fatigue performance index EFPI is less than or equal to the second decision threshold Ar, the design is deemed high-risk, the geometric design is forcibly rejected, and the process returns to step S2 for redesign.

7. The optimization method based on the influence of shear stress amplitude on the cyclic loading characteristics of bonded structures according to claim 6, characterized in that: During the redesign in step S2, the ultimate optimization objective is to maximize the effective fatigue performance index EFPI. The geometric gradient coefficient Kg and the process complexity coefficient Kp are used as design variables. The optimization is carried out iteratively within the preset cost and process capability boundaries to find a balance solution that is not theoretically optimal but has the best overall performance.

8. The optimization method based on the influence of shear stress amplitude on the cyclic loading characteristics of bonded structures according to claim 7, characterized in that: When the maximum effective fatigue performance index EFPI after iterative optimization still cannot exceed the second decision threshold Ar, the system will automatically lock and abandon the complex geometric optimization path and activate the alternative optimization strategy library.

9. The optimization method based on the influence of shear stress amplitude on the cyclic loading characteristics of bonded structures according to claim 8, characterized in that: The alternative optimization strategy library includes: Strategy 1: Keep the original or simplified geometry unchanged, and instead optimize the properties of the adhesive material, such as using functionally graded adhesives or flexible adhesives at the ends; Strategy 2: Based on the original geometric configuration, adopt low-invasive and low-process-risk structural reinforcement measures such as increasing the overlap length or adding external reinforcing patches.

10. The optimization method based on the influence of shear stress amplitude on the cyclic loading characteristics of bonded structures according to claim 9, characterized in that: The method also includes constructing an associated database to store the geometric schemes, process parameters, non-destructive testing schemes and their corresponding calculated values ​​of the geometric stress release index (GSRI), damage invisibility index (DII), and process damage potential index (PDPI) of historical projects, as well as the actual structural life verified by fatigue experiments. The database is trained using a machine learning model to continuously correct the calibration values ​​of the process complexity coefficient Kp and the material damage sensitivity coefficient Km, and to optimize the weighting coefficients, including the allocation strategy of the geometric benefit weighting coefficient, the damage potential weighting coefficient Wp, and the missed detection risk weighting coefficient Wd, thereby improving the accuracy and efficiency of subsequent optimization decisions.

Images