A method and system for predicting fatigue life of a flexible coupling

Abstract

This invention relates to the field of structural fatigue life prediction technology, and discloses a method and system for predicting the fatigue life of a hinged coupling, comprising: Step 1, acquiring torque and input shaft angular position data, mapping time to angular position and dividing it equally to obtain angular domain torque; Step 2, constructing a load vector based on the angular domain torque and mapping matrix, and converting it into bending stress and torsional stress; Step 3, determining the equivalent stress, principal stress surface direction angle, and rotation rate; Step 4, calculating and correcting the average equivalent stress per revolution; Step 5, calculating the load concentration factor based on the assembly misalignment angle and angular position, generating a dual-sector weight core, and combining it with the rotation rate to obtain a follow-up amplification factor; Step 6, integrating the corrected stress, weight core, and amplification factor to obtain the cumulative damage per revolution; Step 7, determining the life by accumulating multi-revolution damage, and estimating the remaining life revolutions and time if no failure occurs. This invention achieves accurate prediction of the fatigue life of hinged couplings under complex working conditions.

Description

Technical Field

[0001] This invention belongs to the field of structural fatigue life prediction technology, specifically relating to a method and system for predicting the fatigue life of a hinged coupling. Background Technology

[0002] Sling couplings, as key connecting components in mechanical transmission systems, are widely used in aerospace, energy equipment, and heavy machinery. Their main function is to compensate for installation misalignments between shafts and transmit power. However, during long-term operation, sling couplings are often subjected to alternating torque and complex multiaxial stresses, making them highly susceptible to fatigue damage. Especially under conditions of assembly misalignment and non-proportional loads, the stress state at the critical section becomes even more complex, making traditional life prediction methods difficult to accurately reflect their true working life.

[0003] Most existing life prediction methods rely on stress or torque signals in the time domain, estimating fatigue life through statistical methods or empirical formulas. However, because time-domain signals are susceptible to rotational speed fluctuations, the sampling points are unevenly distributed within a rotational cycle, thus reducing computational accuracy. Furthermore, existing methods generally lack systematic consideration of assembly errors, non-proportional stress paths, and localized load concentration characteristics, often relying on simplified assumptions to handle stress distribution, failing to comprehensively reflect the fatigue evolution process under complex working conditions. In addition, while finite element simulation can achieve high accuracy, its computational complexity is high, making it unsuitable for real-time or near-real-time life prediction during operation. Summary of the Invention

[0004] This invention provides a method and system for predicting the fatigue life of hinged couplings, solving the technical problems of uneven time-domain data sampling, failure to fully consider assembly misalignment, and non-proportional stress paths that lead to insufficient life prediction accuracy in related technologies.

[0005] This invention provides a method for predicting the fatigue life of a hinged coupling, comprising the following steps:

[0006] Step 1: Obtain the torque time data and input shaft angular position time data passively collected by the hinge coupling, map the time dimension data to the angular position dimension, divide each revolution into M equal division points according to the preset equal division number, and obtain the angular domain torque data of each equal division point;

[0007] Step 2: Based on the angular domain torque data and the pre-pre-fixed stress-load mapping matrix and bending torque coefficient, construct the angular domain load vector and transform the load vector into the bending stress and torsional stress of the critical section.

[0008] Step 3: Based on bending stress and torsional stress, determine the equivalent stress, principal stress surface direction angle, and principal stress surface rotation rate of the critical section;

[0009] Step 4: Calculate the single-rotation average equivalent stress, and correct the equivalent stress based on the average stress and the tensile strength of the material to obtain the corrected equivalent stress.

[0010] Step 5: Calculate the load concentration factor of the corner domain based on the assembly misalignment angle and corner position, and obtain the double sector weight core from it. Then, calculate the follow-up amplification factor based on the principal stress surface rotation rate.

[0011] Step 6: Combining the corrected equivalent stress, dual-sector weighted kernel, and follow-up amplification factor, integrate in the angular domain according to the pre-cured material curve parameters to obtain the single-rotation cumulative damage value;

[0012] Step 7: Accumulate the single-rotation damage values ​​of all observed revolutions to determine the joint coupling; and if it has not failed, calculate the remaining life revolutions and remaining life time.

[0013] Furthermore, using the input shaft angular position time data as a mapping reference, the torque time data is resampled to the angular position dimension. According to the preset number of equal parts, the single revolution is divided into M equal parts, and the torque value of the corresponding time node is extracted at each equal part to obtain the angular domain torque data corresponding to each equal part.

[0014] Furthermore, based on the angular domain torque data and the pre-preserved stress-load mapping matrix and bending torque coefficient, an angular domain load vector is constructed, which is then transformed into the bending and torsional stresses at the critical section, including:

[0015] Step 11: Using the analytical function of assembly misalignment angle and angular position, the angular domain torque data is converted into the corresponding bending torque, and the angular domain torque data is combined with the bending torque to form the angular domain load vector; the analytical function takes the assembly misalignment angle and angular position as input variables and outputs an amplification coefficient that changes periodically with the angular position. This amplification coefficient has two maximum points and two minimum points in one rotation angular domain.

[0016] Step 12: Linearly map the angular domain load vector using the stress-load mapping matrix to obtain the bending stress and torsional stress of the critical section;

[0017] Step 13: Output the bending stress and torsional stress as a normalized angular domain stress dataset. The angular domain stress dataset is a sequence containing M equally divided angular positions, and each equally divided angular position corresponds to a set of bending stress and torsional stress.

[0018] Furthermore, based on bending stress and torsional stress, the equivalent stress, principal stress plane orientation angles, and principal stress plane rotation rates of the critical section are determined, including:

[0019] Step 21: For each equally divided angle position, the energy equivalence criterion is applied to add the square of the bending stress to the square of three times the torsional stress, and the square root of the result is taken to obtain the equivalent stress at that angle position.

[0020] Step 22: For each equally divided angle position, calculate the arctangent of the ratio of twice the torsional stress to the bending stress, and take half of the result to obtain the principal stress surface direction angle at that angle position;

[0021] Step 23: Divide the absolute value of the difference between the principal stress surface direction angles of adjacent equally spaced angle positions by the angle interval to obtain the principal stress surface rotation rate at that angle position.

[0022] Furthermore, the calculation process of the corrected equivalent stress includes:

[0023] Step 31: Within a single revolution range, the equivalent stress at the M equally divided angle positions is arithmetically averaged to obtain the single-revolution average equivalent stress.

[0024] Step 32: Calculate the ratio of the single-rotation average equivalent stress to the material tensile strength to obtain the average stress correction factor;

[0025] Step 33: Divide the equivalent stress value at the angular position by the difference between the value of the average stress correction factor and the value of the equivalent stress value to obtain the corrected equivalent stress.

[0026] Furthermore, the angular domain load concentration factor is calculated based on the assembly misalignment angle and angular position, thereby obtaining the dual-sector weight core, and the follow-up amplification factor is calculated based on the principal stress surface rotation rate, including:

[0027] Step 41: For each equally divided angular position within a single rotation range, calculate the angular domain load concentration factor based on the load concentration factor function of the assembly misalignment angle and the angular position; the load concentration factor function uses a constant as the numerator and a product of the square of the sine of the assembly misalignment angle and the square of the cosine of the angular position as the denominator; the load concentration factor has two symmetrical maximum points within a single rotation angular domain.

[0028] Step 42: Based on the load concentration coefficient of the corner domain, the first-turn corner domain is divided into two high-weight intervals and two low-weight intervals. The high-weight intervals are determined by the two maximum points of the load concentration coefficient, and the low-weight intervals are composed of the remaining corner domains. The normalization process is performed at the M equally divided corner positions by discrete summation so that the sum of the weights in the first-turn range is one, thereby generating a dual-sector weight core.

[0029] Step 43: When the rotation rate of the principal stress surface is less than or equal to the first preset threshold, the follow-up amplification factor is set to one; when the rotation rate of the principal stress surface is greater than the first preset threshold, it increases according to a nonlinear function.

[0030] Furthermore, by combining the corrected equivalent stress, dual-sector weighted kernel, and follow-up amplification factor, and integrating in the angular domain according to the pre-cured material curve parameters, the single-rotation cumulative damage is obtained, including:

[0031] Step 51: Construct a lifetime function based on the preset cured material curve parameters. The lifetime function is a power function relationship between the corrected equivalent stress and the allowable number of cycles. That is, when the stress amplitude is given, calculate the power of the ratio of the corrected equivalent stress to the material fatigue strength constant to obtain the allowable number of cycles; and use the reciprocal of the allowable number of cycles as the damage core.

[0032] Step 52: Input the corrected equivalent stress at each equally divided angle position within a corner domain into the damage kernel, and obtain the first cycle damage value corresponding to the angle position point by point to form an unweighted single-point damage value sequence.

[0033] Step 53: Multiply the unweighted single-point damage value with the corresponding dual-sector weight kernel value and the follow-up amplification factor point by point to obtain the weighted damage value sequence.

[0034] Step 54: Discretely sum the weighted damage value sequence over a turning domain to obtain the cumulative damage value per turn.

[0035] Furthermore, the cumulative damage values ​​per single revolution of all observed revolutions are arithmetically summed to obtain the total cumulative damage value. When the total cumulative damage is greater than or equal to the preset failure threshold, the hinge coupling is determined to have failed due to fatigue. When the total cumulative damage value is less than the preset failure threshold, the hinge coupling is determined not to have failed.

[0036] In the absence of failure, the ratio of the total cumulative damage value to the observed number of revolutions is calculated to obtain the average single-revolution damage value; the result of subtracting the total cumulative damage value from the average single-revolution damage value is divided by the average single-revolution damage value to obtain the remaining lifetime revolutions, and the ratio of the remaining lifetime revolutions to the rotational speed is taken as the remaining lifetime time.

[0037] This invention provides a fatigue life prediction system for hinged couplings, comprising:

[0038] The torque angle mapping module is used to acquire the torque time data and input shaft angular position time data passively collected by the hinge coupling, map the time dimension data to the angular position dimension, divide each revolution into M equal division points according to a preset number of equal division points, and obtain the angular domain torque data of each equal division point;

[0039] The load-stress conversion module is used to construct an angular domain load vector based on angular domain torque data and a pre-pre-built stress-load mapping matrix and bending torque coefficient, and to convert the load vector into bending stress and torsional stress at the critical section.

[0040] The equivalent principal stress calculation module is used to determine the equivalent stress, principal stress surface direction angle, and principal stress surface rotation rate of the critical section based on bending stress and torsional stress.

[0041] The average stress correction module is used to calculate the average equivalent stress per revolution and correct the equivalent stress based on the average stress and the tensile strength of the material to obtain the corrected equivalent stress.

[0042] The weight core amplification factor module is used to calculate the angular domain load concentration factor based on the assembly misalignment angle and angular position, thereby obtaining the dual-sector weight core, and calculating the follow-up amplification factor based on the principal stress surface rotation rate;

[0043] The corner domain integral damage module is used to combine the corrected equivalent stress, dual sector weight kernel and follow-up amplification factor, and integrate in the corner domain according to the pre-cured material curve parameters to obtain the single-rotation cumulative damage value.

[0044] The life determination calculation module is used to accumulate the single-rotation damage value of all observed revolutions, determine the life of the hinge coupling, and calculate the remaining life revolutions and remaining life time if it has not failed.

[0045] The beneficial effects of this invention are as follows: By mapping torque data within the angular domain, this invention eliminates the problem of uneven sampling caused by speed fluctuations in the time domain signal, ensuring the consistency and accuracy of data processing; by introducing an analytical function for assembly misalignment angles and a stress load mapping matrix, it achieves rapid conversion from torque to bending and torsional stresses at critical sections, fully demonstrating the coupling effect of assembly errors and structural stresses; through equivalent stress calculation and average stress correction, combined with material tensile strength parameters, it effectively overcomes the adverse effects of average stress on fatigue life prediction; by constructing a dual-sector weighting core and a follow-up amplification factor, it highlights the characteristics of local load concentration and non-proportional stress paths, thereby improving the sensitivity and accuracy of damage calculation; finally, by combining material fatigue curve parameters for angular domain integration, a life determination and remaining life estimation mechanism is established. Overall, this invention balances computational efficiency and prediction accuracy, improving the fatigue life prediction capability of hinged couplings under complex working conditions. Attached Figure Description

[0046] Figure 1 This is a flowchart of a method for predicting the fatigue life of a hinged coupling according to the present invention. Detailed Implementation

[0047] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, features described in some examples may be combined in other examples.

[0048] like Figure 1 As shown, a method for predicting the fatigue life of a hinged coupling includes the following steps:

[0049] Step 1: Obtain the torque time data and input shaft angular position time data passively collected by the hinge coupling, map the time dimension data to the angular position dimension, divide each revolution into M equal division points according to the preset equal division number, and obtain the angular domain torque data of each equal division point;

[0050] Step 2: Based on the angular domain torque data and the pre-pre-fixed stress-load mapping matrix and bending torque coefficient, construct the angular domain load vector and transform the load vector into the bending stress and torsional stress of the critical section.

[0051] Step 3: Based on bending stress and torsional stress, determine the equivalent stress, principal stress surface direction angle, and principal stress surface rotation rate of the critical section;

[0052] Step 4: Calculate the single-rotation average equivalent stress, and correct the equivalent stress based on the average stress and the tensile strength of the material to obtain the corrected equivalent stress.

[0053] Step 5: Calculate the load concentration factor of the corner domain based on the assembly misalignment angle and corner position, and obtain the double sector weight core from it. Then, calculate the follow-up amplification factor based on the principal stress surface rotation rate.

[0054] Step 6: Combining the corrected equivalent stress, dual-sector weighted kernel, and follow-up amplification factor, integrate in the angular domain according to the pre-cured material curve parameters to obtain the single-rotation cumulative damage value;

[0055] Step 7: Accumulate the single-rotation damage values ​​of all observed revolutions to determine the joint coupling; and if it has not failed, calculate the remaining life revolutions and remaining life time.

[0056] In one embodiment of the present invention, torque-time data and angular position-time data of the input shaft changing over time are first passively acquired by the hinged coupling under natural operating conditions. The torque-time data refers to the transmitted torque value acting on the coupling, recorded sequentially over time by a torque sensor installed at the input end of the coupling or in the transmission system; the input shaft angular position-time data refers to a continuous sequence of the input shaft rotation angle changing over time obtained by an angle encoder or angle sensor, used to reflect the angular displacement of the input shaft throughout the entire rotation process.

[0057] After acquiring the aforementioned time series data, the torque time data is resampled to the angular position dimension using the input shaft angular position time data as a mapping reference. The single revolution is divided into M equal division points according to a preset division ratio, and the torque value at the corresponding time node is extracted at each division point to obtain the angular domain torque data for each division point. Here, the mapping reference refers to using the correspondence between the input shaft angular position and time to transform the original time dimension into an angular dimension, ensuring that subsequent analyses use the rotation angle as a unified reference coordinate.

[0058] Through the above processing, the conversion from time-domain torque data to angular-domain torque data is achieved. This eliminates the non-uniformity of time-domain torque data caused by speed fluctuations, allowing the torque distribution within each revolution range to be sampled uniformly according to the angle, which better reflects the actual stress characteristics of rotating components. Using the angular position as a unified reference dimension facilitates subsequent load conversion, stress analysis, and fatigue life prediction within the angular domain, thereby improving the accuracy and stability of the entire hinge coupling life prediction method.

[0059] In one embodiment of the present invention, an angular domain load vector is constructed based on angular domain torque data and a pre-prepared stress-load mapping matrix and bending torque coefficient. This load vector is then transformed into bending and torsional stresses at the critical section, including:

[0060] Step 11: Using the analytical function of assembly misalignment angle and angular position, the angular torque data is converted into the corresponding bending torque, and the angular torque data is combined with the bending torque to form the angular load vector. The analytical function takes the assembly misalignment angle and angular position as input variables and outputs an amplification factor that changes periodically with the angular position. This amplification factor has two maximum points and two minimum points within a single rotation angular domain. Specifically, the assembly misalignment angle refers to the angular deviation between the input shaft and the output shaft during the assembly process of the coupling, and the angular position refers to the discrete angular points of the input shaft within a single rotation range. The calculation formula of the analytical function is: in, Indicates the magnification factor. This indicates the assembly misalignment angle, where θ represents the angular position. To amplify the bending effect caused by misalignment, the denominator ensures that the amplification factor fluctuates periodically in the angular domain; the bending torque is obtained by multiplying the amplification factor by the angular domain torque data, and then combined with the angular domain torque data to obtain the angular domain load vector.

[0061] Step 12: Linearly map the angular load vector using a stress-load mapping matrix to obtain the bending stress and torsional stress of the critical section. Specifically, the bending stress and torsional stress are obtained by dot multiplying the angular load vector with the stress-load mapping matrix. The stress-load mapping matrix is ​​obtained through finite element analysis and experimental verification during the model finalization stage and remains fixed during operation. Its elements reflect the contribution ratio of different load components to the stress distribution of the critical section. The critical section refers to the structural section of the hinged coupling that is most significantly stressed and most prone to fatigue damage.

[0062] Step 13: Output the bending stress and torsional stress as a normalized angular domain stress dataset. The angular domain stress dataset is a sequence containing M equally divided angular positions, and each equally divided angular position corresponds to a set of bending stress and torsional stress.

[0063] By introducing an analytical function for the assembly misalignment angle, the amplification effect of assembly deviation on bending torque is effectively demonstrated; by linearly transforming the stress load mapping matrix, a rapid conversion from load vector to stress at critical sections is achieved, which reduces computational complexity and avoids the high cost of real-time finite element analysis; by standardizing the output corner domain stress dataset, the integrity and consistency of the data in the corner domain are ensured.

[0064] In one embodiment of the present invention, determining the equivalent stress, principal stress plane orientation angle, and principal stress plane rotation rate of the critical section based on bending stress and torsional stress includes:

[0065] Step 21: For each equally divided angle, the energy equivalence criterion is applied. The square of the bending stress is added to the square of three times the torsional stress, and the square root of the result is taken to obtain the equivalent stress at that angle. The square term ensures the positive cumulative effect of the stress amplitude, and the three-fold coefficient reflects the enhancing effect of shear stress on the equivalent stress in fatigue damage. The equivalent stress refers to the conversion of bending and torsional stresses into a single stress amplitude through the energy equivalence principle when a multi-directional stress state exists, facilitating subsequent fatigue life prediction.

[0066] Step 22: For each equally divided angle, calculate the arctangent of the ratio of twice the torsional stress to the bending stress, and halve the result to obtain the principal stress plane orientation angle at that angle. The orientation angle obtained through the arctangent calculation, and then halved, reflects the symmetry of the principal stress direction under stress conditions. This clarifies the spatial orientation of the principal stress direction on the critical section. The principal stress plane orientation angle refers to the orientation angle of the principal stress direction on the critical section relative to the reference axis.

[0067] Step 23: Divide the absolute value of the difference between the principal stress surface direction angles at adjacent equally spaced angular positions by the angular interval to obtain the principal stress surface rotation rate at that angular position. The principal stress surface rotation rate is used to characterize how quickly the principal stress surface direction changes with angular position, and is an important indicator reflecting the characteristics of non-proportional loads.

[0068] Through the above processing steps, the multiaxial stress state of the critical section can be completely characterized within the angular domain. The introduction of equivalent stress solves the difficulty in comparing bending and torsional stresses due to their different dimensions and directions. The calculation of the principal stress surface direction angles reveals the variation law of the principal stress directions with the angular domain position. The calculation of the principal stress surface rotation rate further quantifies the dynamic characteristics of the principal stress direction changes, providing a basis for subsequently introducing a follow-up amplification factor and correcting for non-proportional fatigue effects.

[0069] In one embodiment of the present invention, the calculation process of the corrected equivalent stress includes:

[0070] Step 31: Within a single rotation range, the equivalent stress at the M equally divided angular positions is arithmetically averaged to obtain the single-rotation average equivalent stress, which is used to eliminate the influence of local fluctuations within the angular domain and to reflect the average stress level within the single rotation range as a whole.

[0071] Step 32: Calculate the ratio of the single-rotation average equivalent stress to the material tensile strength to obtain the average stress correction factor; wherein, the material tensile strength is a parameter determined and solidified through material tensile testing during the model finalization stage;

[0072] Step 33: Divide the equivalent stress value at the angular position by the difference between the average stress correction factor and the value of the average stress correction factor to obtain the corrected equivalent stress. By means of this method, when the average equivalent stress per revolution is high, the average stress correction factor increases accordingly, which improves the corrected equivalent stress and thus reflects the adverse effect of the average equivalent stress per revolution on fatigue damage.

[0073] This embodiment introduces an average stress correction mechanism based on the tensile strength of the material, which can correct the equivalent stress value at the angular domain scale, thereby more realistically reflecting the fatigue stress state of the critical section under actual load conditions.

[0074] In one embodiment of the present invention, the angular domain load concentration factor is calculated based on the assembly misalignment angle and angular position, thereby obtaining the dual-sector weight core, and the follow-up amplification factor is calculated based on the principal stress surface rotation rate, including:

[0075] Step 41: For each equally divided angle position within a single rotation range, calculate the load concentration factor of the angular domain based on the load concentration factor function of the assembly misalignment angle and the angular position. The load concentration factor function uses a constant as the numerator and the product of the square of the sine of the assembly misalignment angle and the square of the cosine of the angular position as the denominator. This load concentration factor has two symmetrical maximum points within a single rotation angle domain. Specifically, the calculation formula for the load concentration factor function is: in, Indicates the angular domain load concentration factor;

[0076] It should be noted that the angular load concentration coefficient function described in this step is not the same as the analytical function in step 11. The analytical function in step 11 is used to convert the additional effect caused by the assembly misalignment angle into bending torque, and the numerator contains a square term to modulate the bending effect. The function in this step, however, uses a constant numerator of 1 and emphasizes the coupling change of the denominator with the assembly misalignment angle and the angular position. It is mainly used to characterize the load concentration degree within the angular domain.

[0077] Step 42: Based on the load concentration coefficient in the corner domain, the first-turn corner domain is divided into two high-weight intervals and two low-weight intervals. The high-weight intervals are determined by the two maximum points of the load concentration coefficient, and the low-weight intervals are composed of the remaining corner domains. The weights are normalized by discrete summation at the M equally divided corner positions so that the sum of the weights within the first-turn range is one, thereby generating a dual-sector weight core. This dual-sector weight core can assign higher weights to fatigue-sensitive areas and lower weights to non-sensitive areas during the corner domain integration process, thereby achieving differentiated processing of damage distribution.

[0078] Step 43: When the rotation rate of the principal stress surface is less than or equal to the first preset threshold, the follow-up amplification factor is set to one; when the rotation rate of the principal stress surface is greater than the first preset threshold, it increases according to a nonlinear function.

[0079] This embodiment reveals the load concentration effect caused by assembly misalignment angle through a load concentration factor function; it assigns differentiated weights to fatigue-sensitive and non-sensitive areas through a dual-sector weighting kernel; and it enhances the damage effect in the rapidly changing principal stress direction range through a follow-up amplification factor. Overall, it can improve the adaptability and prediction accuracy of the fatigue life prediction method for hinged couplings to complex load conditions.

[0080] In one embodiment of the present invention, by combining the corrected equivalent stress, dual-sector weighted kernel, and follow-up amplification factor, and integrating in the angular domain according to the pre-cured material curve parameters, the single-rotation cumulative damage is obtained, including:

[0081] Step 51: Construct a lifetime function based on the preset cured material curve parameters. The lifetime function is a power function relationship between the corrected equivalent stress and the allowable number of cycles. That is, for a given stress amplitude, calculate the power of the ratio of the corrected equivalent stress to the material fatigue strength constant to obtain the allowable number of cycles; and use the reciprocal of the allowable number of cycles as the damage nucleus. Specifically, the material curve parameters refer to the material fatigue strength constant and material fatigue index obtained and cured through standard fatigue tests during the model finalization stage, which are used to establish the relationship between the corrected equivalent stress and fatigue life. The calculation formula for the lifetime function is:

[0082] N f Indicates the allowed number of loops. The value represents the corrected equivalent stress, σ0 represents the material fatigue strength constant, and m represents the material fatigue index; the damage core is used to represent the amount of fatigue damage corresponding to a single cycle under a given stress amplitude.

[0083] Step 52: Input the corrected equivalent stress at each equally divided angle position within a corner domain into the damage kernel, and obtain the first cycle damage value corresponding to the angle position point by point to form an unweighted single-point damage value sequence.

[0084] Step 53: Multiply the unweighted single-point damage value with the corresponding angular position dual-sector weight core value and the follow-up amplification factor point by point to obtain the weighted damage value sequence; the follow-up amplification factor is used to enhance the fatigue effect when the principal stress direction changes rapidly.

[0085] Step 54: Discretely sum the weighted damage value sequence over a turning domain to obtain the cumulative damage value per turn.

[0086] This embodiment achieves multiple corrections for load distribution and fatigue effects through an integration method within the angular domain; by introducing a dual-sector weighting kernel and a follow-up amplification factor, it ensures the sensitivity of damage calculation to local load concentration and non-proportional stress; and by solidifying the use of material curve parameters, it ensures the stability and repeatability of the prediction results. Overall, this embodiment improves the accuracy and engineering applicability of fatigue life prediction for hinged couplings.

[0087] In one embodiment of the present invention, the cumulative damage values ​​of each single revolution of all observed revolutions are arithmetically summed to obtain the total cumulative damage value. When the total cumulative damage is greater than or equal to a preset failure threshold, the hinge coupling is determined to have suffered fatigue failure. When the total cumulative damage value is less than the preset failure threshold, the hinge coupling is determined not to have failed. Preferably, the preset failure threshold is set to 1.

[0088] In the absence of failure, the ratio of the total cumulative damage value to the observed number of revolutions is calculated to obtain the average single-revolution damage value; the result of subtracting the total cumulative damage value from the average single-revolution damage value is divided by the average single-revolution damage value to obtain the remaining lifetime revolutions, and the ratio of the remaining lifetime revolutions to the rotational speed is taken as the remaining lifetime time.

[0089] This embodiment ensures that fatigue assessment can reflect the observed operating status in real time by introducing the total cumulative damage value; the life estimation method based on the average single-rotation damage value can quickly estimate the remaining life rotations and remaining life time in the absence of failure, thereby providing quantifiable life indicators for operation and maintenance, which helps to carry out preventive maintenance or replacement of the joint coupling in advance.

[0090] This embodiment also provides a fatigue life prediction system for hinged couplings, including:

[0091] The torque angle mapping module is used to acquire the torque time data and input shaft angular position time data passively collected by the hinge coupling, map the time dimension data to the angular position dimension, divide each revolution into M equal division points according to a preset number of equal division points, and obtain the angular domain torque data of each equal division point;

[0092] The load-stress conversion module is used to construct an angular domain load vector based on angular domain torque data and a pre-pre-built stress-load mapping matrix and bending torque coefficient, and to convert the load vector into bending stress and torsional stress at the critical section.

[0093] The equivalent principal stress calculation module is used to determine the equivalent stress, principal stress surface direction angle, and principal stress surface rotation rate of the critical section based on bending stress and torsional stress.

[0094] The average stress correction module is used to calculate the average equivalent stress per revolution and correct the equivalent stress based on the average stress and the tensile strength of the material to obtain the corrected equivalent stress.

[0095] The weight core amplification factor module is used to calculate the angular domain load concentration factor based on the assembly misalignment angle and angular position, thereby obtaining the dual-sector weight core, and calculating the follow-up amplification factor based on the principal stress surface rotation rate;

[0096] The corner domain integral damage module is used to combine the corrected equivalent stress, dual sector weight kernel and follow-up amplification factor, and integrate in the corner domain according to the pre-cured material curve parameters to obtain the single-rotation cumulative damage value.

[0097] The life determination calculation module is used to accumulate the single-rotation damage value of all observed revolutions, determine the life of the hinge coupling, and calculate the remaining life revolutions and remaining life time if it has not failed.

[0098] It should be noted that the interval and threshold sizes are set for ease of comparison. The size of the threshold depends on the amount of sample data and the base number set by those skilled in the art for each set of sample data, as long as it does not affect the proportional relationship between the parameter and the quantized value. Furthermore, the above formulas are all dimensionless calculations, and the formulas are derived from software simulations using a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0099] The embodiments of the present invention have been described above. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms based on the guidance of the present embodiments, all of which are within the protection scope of the present embodiments.

Claims

1. A method for predicting the fatigue life of a hinged coupling, characterized in that, Includes the following steps: Step 1: Obtain the torque time data and input shaft angular position time data passively collected by the hinge coupling, map the time dimension data to the angular position dimension, divide each revolution into M equal division points according to the preset equal division number, and obtain the angular domain torque data of each equal division point; Step 2: Based on the angular domain torque data and the pre-pre-fixed stress-load mapping matrix and bending torque coefficient, construct the angular domain load vector and transform the load vector into the bending stress and torsional stress of the critical section. Step 3: Based on bending stress and torsional stress, determine the equivalent stress, principal stress surface direction angle, and principal stress surface rotation rate of the critical section; Step 4: Calculate the single-rotation average equivalent stress, and correct the equivalent stress based on the average stress and the material tensile strength to obtain the corrected equivalent stress, including: Within a single revolution, the equivalent stress at M equally divided angle positions is arithmetically averaged to obtain the single-revolution average equivalent stress. The average stress correction factor is obtained by calculating the ratio of the single-rotation average equivalent stress to the tensile strength of the material. Divide the equivalent stress value at the angular position by the difference between the value of the equivalent stress value and the value of the mean stress correction factor to obtain the corrected equivalent stress. Step 5: Calculate the angular load concentration factor based on the assembly misalignment angle and angular position, thereby obtaining the dual-sector weight core, and calculate the follow-up amplification factor based on the principal stress surface rotation rate, including: For each equally divided angular position within a single rotation range, the angular domain load concentration factor is calculated based on the load concentration factor function of the assembly misalignment angle and the angular position. The load concentration factor function uses a constant as the numerator and a product of the square of the sine of the assembly misalignment angle and the square of the cosine of the angular position as the denominator. This load concentration factor has two symmetrical maximum points within a single rotation angular domain. Based on the load concentration coefficient of the corner domain, the first-turn corner domain is divided into two high-weight intervals and two low-weight intervals. The high-weight intervals are determined by the two maximum points of the load concentration coefficient, and the low-weight intervals are composed of the remaining corner domains. The normalization process is performed at the M equally divided corner positions by discrete summation so that the sum of the weights within the first-turn range is one, thereby generating a dual-sector weight core. When the rotation rate of the principal stress surface is less than or equal to the first preset threshold, the follow-up amplification factor is set to one; when the rotation rate of the principal stress surface is greater than the first preset threshold, it increases according to a nonlinear function. Step 6: Combining the corrected equivalent stress, dual-sector weighted kernel, and follow-up amplification factor, integrate in the angular domain according to the pre-cured material curve parameters to obtain the single-rotation cumulative damage value; Step 7: Accumulate the single-rotation damage values ​​of all observed revolutions to determine the joint coupling; and if it has not failed, calculate the remaining life revolutions and remaining life time.

2. The fatigue life prediction method for a hinged coupling according to claim 1, characterized in that, Using the input shaft angular position time data as a mapping reference, the torque time data is resampled to the angular position dimension. According to the preset number of equal parts, the single revolution is divided into M equal parts, and the torque value of the corresponding time node is extracted at each equal part, so as to obtain the angular domain torque data corresponding to each equal part.

3. The fatigue life prediction method for a hinged coupling according to claim 1, characterized in that, Based on angular domain torque data and pre-preserved stress-load mapping matrix and bending torque coefficient, an angular domain load vector is constructed, which is then transformed into bending and torsional stresses at the critical section, including: Using analytical functions of assembly misalignment angle and angular position, the angular torque data is converted into the corresponding bending torque, and the angular torque data is combined with the bending torque to form an angular load vector; the analytical function takes assembly misalignment angle and angular position as input variables and outputs an amplification factor that changes periodically with the angular position. This amplification factor has two maximum points and two minimum points in one rotation angular domain. The bending stress and torsional stress of the critical section are obtained by linearly mapping the angular domain load vector using a stress-load mapping matrix. The bending stress and torsional stress are output as a normalized angular domain stress dataset, which is a sequence containing M equally divided angular positions, each of which corresponds to a set of bending stress and torsional stress.

4. The fatigue life prediction method for a hinged coupling according to claim 1, characterized in that, Based on bending and torsional stresses, the equivalent stress, principal stress plane orientation angles, and principal stress plane rotation rates of the critical section are determined, including: For each equally divided angle, the energy equivalence criterion is applied, the square of the bending stress is added to the square of three times the torsional stress, and the square root of the result is taken to obtain the equivalent stress at that angle. For each equally divided angle position, calculate the arctangent of the ratio of twice the torsional stress to the bending stress, and take half of the result to obtain the principal stress surface direction angle at that angle position; The rotation rate of the principal stress surface at that angle is obtained by dividing the absolute value of the difference between the direction angles of the principal stress surfaces at adjacent equally spaced angles by the angle interval.

5. The fatigue life prediction method for a hinged coupling according to claim 1, characterized in that, Combining the corrected equivalent stress, dual-sector weighted kernel, and follow-up amplification factor, the cumulative damage per revolution is obtained by integrating in the angular domain according to the pre-cured material curve parameters, including: A lifetime function is constructed based on the preset cured material curve parameters. The lifetime function is a power function relationship between the corrected equivalent stress and the allowable number of cycles. That is, when the stress amplitude is given, the power of the ratio of the corrected equivalent stress to the material fatigue strength constant is calculated to obtain the allowable number of cycles; and the reciprocal of the allowable number of cycles is used as the damage core. The corrected equivalent stress at each equally divided angle position within a turning angle domain is input into the damage kernel, and the first cycle damage value corresponding to that angle position is obtained point by point, forming an unweighted single-point damage value sequence. The unweighted single-point damage value is multiplied point by point with the corresponding dual-sector weight core value and the follow-up amplification factor to obtain the weighted damage value sequence; Discretely sum the weighted damage value sequence over a single-turn domain to obtain the cumulative damage value per turn.

6. The fatigue life prediction method for a hinged coupling according to claim 1, characterized in that, The cumulative damage values ​​of each observed revolution are summed arithmetically to obtain the total cumulative damage value. When the total cumulative damage is greater than or equal to the preset failure threshold, the hinge coupling is determined to have failed due to fatigue. When the total cumulative damage is less than the preset failure threshold, the hinge coupling is determined not to have failed. In the absence of failure, the ratio of the total cumulative damage value to the observed number of revolutions is calculated to obtain the average single-revolution damage value; the result of subtracting the total cumulative damage value from the average single-revolution damage value is divided by the average single-revolution damage value to obtain the remaining lifetime revolutions, and the ratio of the remaining lifetime revolutions to the rotational speed is taken as the remaining lifetime time.

7. A fatigue life prediction system for hinged couplings, characterized in that, The fatigue life prediction method for a hinged coupling as described in any one of claims 1-6 includes: The torque angle mapping module is used to acquire the torque time data and input shaft angular position time data passively collected by the hinge coupling, map the time dimension data to the angular position dimension, divide each revolution into M equal division points according to a preset number of equal division points, and obtain the angular domain torque data of each equal division point; The load-stress conversion module is used to construct an angular domain load vector based on angular domain torque data and a pre-pre-built stress-load mapping matrix and bending torque coefficient, and to convert the load vector into bending stress and torsional stress at the critical section. The equivalent principal stress calculation module is used to determine the equivalent stress, principal stress surface direction angle, and principal stress surface rotation rate of the critical section based on bending stress and torsional stress. The average stress correction module is used to calculate the average equivalent stress per revolution and correct the equivalent stress based on the average stress and the tensile strength of the material to obtain the corrected equivalent stress. The weight core amplification factor module is used to calculate the angular domain load concentration factor based on the assembly misalignment angle and angular position, thereby obtaining the dual-sector weight core, and calculating the follow-up amplification factor based on the principal stress surface rotation rate; The corner domain integral damage module is used to combine the corrected equivalent stress, dual sector weight kernel and follow-up amplification factor, and integrate in the corner domain according to the pre-cured material curve parameters to obtain the single-rotation cumulative damage value. The life determination calculation module is used to accumulate the single-rotation damage value of all observed revolutions, determine the life of the hinge coupling, and calculate the remaining life revolutions and remaining life time if it has not failed.

Images