Rudder assembly performance degradation analysis method

Abstract

This invention relates to the field of mechanical structure performance testing technology, and discloses a method for analyzing the performance degradation of a rudder assembly to improve the accuracy of multi-component coupled degradation analysis without disassembly. The method includes: acquiring time-series monitoring data of the torsion spring deployment driving force, fitting and outputting its time-series decay variable over the storage period; simulating the random fluctuations and nonlinear degradation laws of the resistance of the articulated joint and locking mechanism over the storage period based on the degradation mechanism of internal resistance elements; calculating the actual effective deployment driving force by using real-time driving force to gradually offset the frictional resistance of the articulated joint and the jamming resistance of the locking mechanism according to the mechanical transmission logic; and extracting single-index degradation curves for mechanical margin, response speed, and positioning reliability by combining the driving force decay and resistance degradation laws, and generating a comprehensive degradation trajectory of the rudder assembly deployment performance through correlation fusion.

Description

Technical Field

[0001] This invention relates to the field of mechanical structure performance testing technology, and in particular to a method for analyzing the performance degradation of a rudder assembly. Background Technology

[0002] The rudder assembly is the core actuator for attitude control of various aircraft and missiles. Its deployment reliability and driving force stability after long-term storage directly determine the overall operational safety and mission success rate. The rudder assembly is an integrated precision structure, mainly composed of multiple key components such as torsion spring drive structure, articulated transmission pair, and locking and limiting mechanism to complete the deployment action. Its overall performance degradation is not caused by the deterioration of a single component, but is the comprehensive result of the coordinated deterioration of multiple components.

[0003] The articulated joint is mounted at the rotational engagement position between the control surface and the control shaft, and between the control shaft and the fuselage mounting base, providing rotational support and constraint for the rudder assembly's retraction and deployment movements. The locking mechanism, coaxially arranged adjacent to the articulated joint and integrated within the rotational engagement cavity of the control assembly, is used to lock and fix the control surface in its folded storage state and rigidly limit its deployment. The torsion spring, as the power output element for the rudder assembly's deployment movement, is fitted at the hinged rotation position of the control shaft, forming a mechanical relationship with both the articulated joint and the locking mechanism: the torsion spring stores elastic potential energy and provides the driving force for control surface deployment; the articulated joint generates frictional damping resistance during the torsion spring's rotational movement; and the locking mechanism generates jamming resistance and engagement frictional resistance during unlocking. During long-term storage of the rudder assembly, the attenuation of the torsion spring's elastic preload and the damping degradation of the articulated joint and locking mechanism are coupled together, causing a decline in the overall deployment force performance of the rudder assembly. This is the core structural factor affecting the accuracy of the rudder assembly's storage life assessment.

[0004] Therefore, in the process of analyzing the storage performance degradation of the rudder assembly, it is theoretically necessary to combine the synchronous degradation characteristics of multiple key components of the rudder assembly to carry out coupled analysis in order to truly and completely reflect the actual degradation law of the entire rudder assembly.

[0005] However, in practical engineering applications, the rudder assembly is a sealed, integrated precision component. After leaving the factory, it is in a state of complete assembly and sealed storage for a long time, which presents objective limitations such as the inability to disassemble for testing and the high requirements for structural assembly precision. If the internal components are forcibly disassembled for testing, it is very easy to damage the original assembly precision, structural fit, and sealed storage state of the rudder assembly, causing the product under test to be scrapped directly. The testing cost is extremely high and it is not feasible for batch engineering testing. Due to this technical difficulty, existing technologies cannot directly carry out long-term in-situ degradation monitoring of each key internal component of the rudder assembly, cannot obtain real time-series degradation data of multiple components, and therefore cannot achieve multi-component coupled whole-machine performance degradation analysis.

[0006] To address the issue of not being able to perform actual disassembly and testing, existing technologies generally employ simulation to predict the storage performance of rudder components. However, traditional simulation methods can only perform theoretical deductions based on preset ideal structural parameters and fixed boundary conditions, resulting in significant technical limitations: First, simulation models cannot realistically reproduce the random structural degradation characteristics brought about by long-term storage environments; second, traditional simulations often use single-factor analysis modes, considering only the attenuation of a single structure and ignoring the coupled degradation effects of multiple structures, leading to a large deviation from the collaborative degradation mechanism of multiple components under actual operating conditions; third, simulation results are theoretical deduction data, not based on actual measured data of rudder component storage degradation, resulting in distorted final performance degradation analysis results, low accuracy in lifespan prediction, and poor reliability.

[0007] In summary, current rudder component performance degradation analysis technologies suffer from several problems: difficulty in implementing disassembly and actual testing, challenges in implementing multi-component coupled analysis, and the use of simulation methods suffers from distorted performance degradation analysis results, low accuracy in lifespan prediction, and poor reliability. Summary of the Invention

[0008] The purpose of this invention is to disclose a method for analyzing the performance degradation of a rudder assembly, so as to improve the accuracy of multi-component coupling degradation analysis without disassembling the machine.

[0009] To achieve the above objectives, the present invention discloses a method for analyzing the performance degradation of a rudder assembly, comprising the following steps:

[0010] Step S10: Obtain the timing monitoring data of the torsion spring deployment driving force of the rudder assembly within the preset storage period. Based on the timing monitoring data of the torsion spring deployment driving force, fit the timing decay variable of the torsion spring deployment driving force with the storage period. Step S20: Based on the degradation mechanism of each resistance element inside the rudder assembly, simulate the degradation law of each resistance element inside the rudder assembly with the storage period; wherein, the resistance element inside the rudder assembly includes the articulated joint and the locking mechanism, and the degradation mechanism of the resistance element is the random fluctuation and nonlinear degradation characteristics of the resistance caused by the performance change of the articulated joint and the locking mechanism under the long-term storage environment. Step S30: Based on the mechanical transmission logic of the rudder assembly, the remaining driving force after the real-time torsion spring deployment driving force offsets the real-time articulated pair friction resistance and the real-time locking mechanism jamming resistance is used to simulate the actual effective deployment driving force of the rudder assembly. Step S40: Combining the time-series decay variable of the torsion spring deployment driving force with the degradation law of the internal resistance element of the rudder assembly, extract the single-index degradation curves corresponding to the time-series sequences of mechanical margin, response speed and positioning reliability, and perform correlation fusion on each single-index degradation curve to form the comprehensive degradation trajectory of the rudder assembly deployment performance.

[0011] Preferably, step S10 includes: Step S11: Determine the complete storage cycle based on the preset storage life of the rudder component and set the sampling nodes for time-series monitoring; Step S12: At each sampling node, the in-situ multiple rounds of retesting and collection of the deployment driving force corresponding to the built-in torsion spring of the rudder assembly are performed to obtain multiple sets of original mechanical test data. Step S13: Preprocess the original mechanical test data, remove failed and distorted data by combining the preset normal performance range of the rudder component, and complete the data noise reduction optimization by selecting the best of multiple sets of retest data. Step S14: Correlate the preprocessed effective deployment driving force data with the corresponding storage reduction time in a time series to construct a continuous time series monitoring dataset that forms a time-mechanical performance correspondence. Step S15: Extract the completed continuous time-series monitoring dataset, and build a basic data sample set for time-series fitting with storage time as the independent variable and the preprocessed torsion spring unfolding driving force as the dependent variable. Step S16: Determine the attenuation fitting method by combining the stress relaxation and constitutive characteristics of the torsion spring metal material during long-term storage and aging, and determine the initial values ​​of the basic parameters of the attenuation fitting method. Step S17: Substitute the continuous time-series monitoring dataset into the attenuation fitting method and calibrate the attenuation fitting method. Step S18: Based on the calibrated attenuation fitting method, output the dynamic time-series attenuation variable of the continuous torsion spring deployment driving force with storage time, and obtain the real-time attenuation value of the deployment driving force and the attenuation rate change law during the entire storage period.

[0012] Preferably, step S20 includes: Step S21: Retrieve the time-series environmental operating condition parameters within the preset storage period, construct the correlation model of storage environment operating condition parameters-structural degradation-resistance increment, and obtain the deterioration coefficient of the hinge pair and the deterioration coefficient of the locking mechanism based on the time-series environmental operating condition parameters. Step S22: Substitute the deterioration coefficient of the articulated joint into the nonlinear quantitative relationship between the cumulative deterioration and the frictional resistance, and combine it with the sampling nodes that are synchronized with the attenuation time of the torsion spring's unfolding driving force to determine the basic increment of the frictional resistance of the articulated joint and the amount of environmental random fluctuation at each sampling node. Step S23: Update the dynamic fluctuation range of the dynamic jamming resistance based on the deterioration coefficient of the locking mechanism. Combined with the irreversible deterioration mechanism of the locking mechanism, introduce a random mutation coefficient to simulate and generate the instantaneous abnormal peak value of the jamming resistance. Step S24: Integrate the basic increment of the frictional resistance of the articulated pair and the random fluctuation of the environment under each sampling node, as well as the dynamic fluctuation range and instantaneous abnormal peak value of the dynamic jamming resistance of the locking mechanism, to obtain the deterioration time sequence curve of the frictional resistance of the articulated pair and the deterioration time sequence curve of the jamming resistance of the locking mechanism, so as to realize the time-series synchronous deduction of the structural resistance deterioration process and the torsion spring deployment driving force deterioration process.

[0013] Preferably, step S30 includes: Step S31: Organize the step-by-step mechanical transmission path of the rudder assembly to construct the mechanical priority logic of torsion spring deployment driving force, overcoming the frictional resistance of the articulated pair and overcoming the locking resistance to the effective driving force output. Step S32: Time-match the real-time torsion spring deployment driving force, real-time hinge pair friction resistance, and real-time locking mechanism jamming resistance of the same sampling node. Step S33: According to the mechanical transmission path, perform mechanical difference calculation step by step to offset the basic increment of frictional resistance of the articulated pair and the random fluctuation of the environment, as well as the dynamic fluctuation range and instantaneous abnormal peak value of dynamic jamming resistance of the locking mechanism. Step S34: Iteratively update the effective expansion driving force values ​​of each sampling node throughout the entire storage cycle in real time to form a continuous time-series change dataset of driving force deteriorating bidirectionally over time.

[0014] Preferably, step S40 includes: Step S41: Extract the core degradation variables, including the time-series decay variable of the torsion spring deployment driving force, the degradation law of the frictional resistance of the hinge pair with the storage period, and the degradation law of the jamming resistance of the locking mechanism with the storage period. Step S42: Extract the differential degradation features corresponding to the time series of mechanical margin, response speed and on-time reliability, and perform time series fitting to construct the corresponding single-index degradation curve; Step S43: Perform correlation fusion on the single-index degradation curves, correct the degradation coupling deviation between multiple indices, and form a unified comprehensive degradation trajectory of the rudder component's deployment performance.

[0015] Preferably, step S40 further includes: Step S44: By identifying critical decay nodes and abrupt deterioration nodes in the integrated degradation trajectory, the complete evolution process of the rudder component performance from normal, weakened to critical failure is characterized.

[0016] Preferably, the method of the present invention further includes: Step S60: Pre-store the standard failure thresholds for the mechanical margin, response speed, and positioning reliability of the rudder assembly, and then match the performance degradation trajectory of the rudder assembly with the failure thresholds. Step S70: Use the trajectory extrapolation method to estimate the time node when the performance of the rudder component reaches the failure threshold, so as to predict the remaining reliable storage life of the rudder component.

[0017] Furthermore, the method of the present invention also includes: Step S80: Quantify and calculate the contribution of torsion spring preload degradation and internal drag element degradation to the performance degradation of the rudder assembly, respectively. Step S90: Based on the proportion of degradation contribution, distinguish between core failure causes and secondary influencing factors, and complete the storage reliability classification evaluation and failure risk location of the rudder component.

[0018] The method of the present invention has at least the following beneficial effects: Firstly, this invention is based on the timing monitoring data of the torsion spring deployment driving force that can be collected in situ. The entire process does not require disassembly of the rudder assembly, and non-destructive storage degradation analysis can be completed without damaging the product structure and storage status. This effectively reduces testing costs and is suitable for engineering scenarios of long-term storage reliability assessment of batch products, making it more practical and widely applicable.

[0019] Secondly, this invention uses measured data on the degradation of the driving force as its core, and combines it with the actual storage degradation mechanism of the internal resistance components of the rudder assembly to simultaneously deduce the collaborative degradation law of the articulated joint and the locking mechanism, thereby reproducing the unsteady degradation characteristics of multiple resistance components. This effectively makes up for the technical defect of traditional simulation results being out of sync with actual working conditions.

[0020] Furthermore, this invention overcomes the limitations of single-factor degradation analysis that only analyzes the torsion spring of the rudder assembly, and realizes a comprehensive performance degradation trajectory through dual-factor coupled collaborative analysis of torsion spring preload degradation and structural resistance degradation, thereby improving the accuracy of multi-component coupled degradation analysis without disassembly.

[0021] The present invention will now be described in further detail with reference to the accompanying drawings. Attached Figure Description

[0022] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a flowchart of the rudder component performance degradation analysis method disclosed in the embodiments of the present invention.

[0023] Figure 2 This is the equivalent storage life of the high-temperature accelerated test disclosed in the embodiments of the present invention - unfolded force diagram.

[0024] Figure 3 This is an equivalent storage life-expanding force diagram of the high temperature and high humidity accelerated test disclosed in the embodiments of the present invention.

[0025] Figure 4 It is the "dense time-driving force" curve of least squares linear regression obtained by integrating six groups of samples as disclosed in the embodiments of the present invention. Detailed Implementation

[0026] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings, but the present invention can be implemented in many different ways as defined and covered by the claims.

[0027] Example 1 Please see Figures 1 to 4 This embodiment discloses a method for analyzing the performance degradation of a rudder assembly, including the following steps: Step S10: Obtain the timing monitoring data of the torsion spring deployment driving force of the rudder assembly within a preset storage period. Based on the timing monitoring data of the torsion spring deployment driving force, fit the timing decay variable of the torsion spring deployment driving force with the storage period.

[0028] This step allows for in-situ detection of the static deployment driving force of the torsion spring at each preset sampling node without disassembling the rudder assembly housing or damaging the original assembly structure, collecting multiple sets of raw measured data. The effective deployment driving force values ​​are then correlated with corresponding storage durations to form a continuous time-series dataset of the torsion spring deployment driving force. Considering the material properties of the spring steel used in the torsion spring, including long-term storage stress relaxation and natural aging, a nonlinear decay fitting method adapted to the monotonic gradual decrease variation law is selected. After parameter iteration calibration and error verification, a continuous decay curve of the torsion spring deployment driving force throughout the entire cycle is output, clearly defining the deployment driving force value and real-time decay rate at any storage time. Specifically, this may include: Step S11: Determine the complete storage cycle based on the preset storage life of the rudder component and set the sampling nodes for time-series monitoring.

[0029] In step S12, at each sampling node, the in-situ multiple rounds of re-measurement are performed on the deployment driving force corresponding to the built-in torsion spring of the rudder assembly to obtain multiple sets of original mechanical test data.

[0030] Step S13: Preprocess the original mechanical test data, remove failed and distorted data by combining the preset normal performance range of the rudder component, and complete the data noise reduction optimization by selecting the best of multiple sets of retest data (e.g., the minimum or average value is preferred).

[0031] Step S14: The preprocessed effective deployment driving force data is correlated with the corresponding storage reduction time (also known as the reduced time or equivalent acceleration time) in a time series to construct a continuous time series monitoring dataset that forms a time-mechanical performance correspondence.

[0032] During service, the mechanical properties of rudder components are easily degraded by environmental factors such as temperature and humidity, directly affecting their reliability and lifespan. To quickly assess the performance degradation patterns of rudder components, high-temperature and high-temperature-high-humidity accelerated tests were conducted on rudder components of the same model. These accelerated tests effectively simulated a long-term natural storage environment, collecting data on the changes in the deployment driving force of the products under different simplification periods.

[0033] (1) Test type: Two test environments are set up: 85℃ high temperature accelerated condition and 85℃, 85%RH high temperature and high humidity accelerated condition.

[0034] (2) Test parameters: The independent variable is the reduced time X (equivalent storage years), and the dependent variable is the unfolding driving force Y, in N.

[0035] (3) Test method: The measurement was performed on 11 samples with 3 different numbers. The test data was collected one day after the samples were placed. To ensure the safety performance of the rudder assembly, the minimum or average value of multiple measurements taken at the same time was used as the reference value for any sample. For example, the distribution of the 3 valid measurements was as follows: the fully retracted sample was slowly released to naturally expand to the maximum stroke, and the first expansion force value was read and recorded; then the sample was retracted 5 times, and the above expansion measurement steps were repeated 2 times, for a total of 3 valid measurements. The average time of the expansion action was recorded simultaneously when the 3 valid measurements were completed (to provide data samples for the subsequent step S42 response speed timing fitting).

[0036] (4) Normal performance range: Stress values ​​below a set threshold are judged as performance degradation to failure state. For example, stress values ​​of 10.5N to 17N, below 10.5N are judged as performance degradation to failure state. Failure data are still used in subsequent fitting.

[0037] After the experiments were completed, the accelerated test data at 85℃ included three sets of samples (Y1, Y2, Y3), corresponding to 12 reduced time points (0, 0.25, 0.5, ..., 16); the accelerated test data at high temperature and high humidity also included three sets of samples (Y4, Y5, Y6), corresponding to 10 reduced time points (0, 0.5, 1, ..., 16). (Refer to...) Figure 2 and Figure 3 In this context, X represents the equivalent storage life and Y represents the unfolding force (N).

[0038] During the experiment, environmental factors were the key factors affecting the storage life of the torsion spring of the rudder assembly. Among them, the synergistic effect of temperature and humidity accelerated the stress relaxation process of the material. Figure 2 and Figure 3The accelerated testing results showed significant differences in the performance degradation rate and failure time among different samples. Samples Y1 and Y4 experienced performance values ​​dropping below the normal range within a short testing period, with an equivalent storage life of less than one year; sample Y2 failed within a medium testing period, with an equivalent storage life of four years; sample Y5 failed in a later testing period, with an equivalent storage life of 12 years; samples Y3 and Y6 maintained performance values ​​within the normal range throughout the entire testing cycle, with an effective storage life of over 16 years. Under initial high stress conditions, temperature or humidity became the dominant influencing factor, accelerating material stress relaxation, driving the continuous decay of torsion spring preload, and ultimately leading to failure problems such as insufficient control surface deployment torque. Individual differences in production quality among different samples, combined with environmental factors, also resulted in significant differentiation in lifespan performance. This pattern also verifies the performance degradation conformal function characteristics, providing a reliable basis for lifespan assessment.

[0039] Linear interpolation was performed on the original data. The high-temperature test data (Data1) was interpolated to generate a dataset Data3 with 65 reduced time points (0~16, step size 0.25), and the high-temperature and high-humidity test data (Data2) was interpolated to generate a dataset Data4 with the same 65 time points. The interpolation rule was: if the interpolated time point coincided with the original data time point, the original stress value was directly used; if they did not coincide, the interpolation result was obtained by linearly calculating the stress values ​​of adjacent time points.

[0040] Datasets Data3 and Data4 were merged to form a comprehensive dataset Data5 containing 6 groups of samples (Y1-Y6) and 65 time points.

[0041] A fitting method is used to fit the data to six sets of samples, and the fitting results serve as the dedicated time-series data for the rudder component. For example, in this embodiment, the six sets of sample data from Data5 are used to construct the gray sequence Data6 using an accumulation generation method, enhancing the regularity of the data and reducing the influence of random errors. The parameters of the discrete gray system are then solved using the least squares method. and The linear regression equation is: ; Where i is the sample group number, used to distinguish multiple groups of stress test samples with parallel performance degradation. In this embodiment, there are a total of 6 groups of samples. For the current discrete time-series node, =1,2,3,...n, where n is the total number of observation points in a single sequence; The next discrete observation time series node to be predicted; Generate a gray sequence value for the first accumulation of the i-th sample and the k-th discrete time point; The predicted output value of the cumulative gray sequence corresponding to the i-th sample and the (k+1)-th discrete time point is given. The slope of the linear regression equation represents the overall rate of change in the performance degradation data. The intercept of the linear regression equation represents the steady-state driving term of the system.

[0042] Solve using the least squares method , Two model parameters: The gray sequence corresponding to =1,2,3,...,n As the independent variable X, =1,2,3,...,n corresponds to The design matrix is ​​constructed using Y as the dependent variable. The regression coefficients are solved by left division of the matrix, and the goodness of fit of each group of samples is calculated. It can accurately capture the temporal variation law of equipment stress degradation; and solve for... , Then, using the conversion formula: , The development coefficients of the traditional GM(1,1) model are obtained. ash action amount .

[0043] Please refer to Figure 4 , Figure 4 To synthesize the least squares linear regression curves obtained from the six sets of data samples, continuous time-series monitoring data on the relationship between storage time and deployment driving force were obtained by fitting the sampled data of each rudder component. The continuous time-series monitoring data corresponding to each time point were used to construct a continuous time-series monitoring dataset.

[0044] In other words, by following the steps above, a continuous degradation curve can be calculated using the grey model. Then, the values ​​at each time point on this curve can be extracted and organized into a "dense time-driving force" data table containing all the predicted values.

[0045] Based on the expansion of sparse measured points into dense time-series data completed in steps S11 to S14 above, in order to upgrade the "mathematical prediction result" of the grey model to an "engineering degradation model that conforms to the physical aging law", step S10 further includes: Step S15: Extract the completed continuous time-series monitoring dataset, and construct a basic data sample set for time-series fitting with storage time as the independent variable and the preprocessed torsion spring unfolding driving force as the dependent variable.

[0046] Step S16: Determine the attenuation fitting method based on the stress relaxation and constitutive characteristics of the torsion spring metal material during long-term storage, and determine the initial values ​​of the basic parameters of the attenuation fitting method. For example, the initial values ​​of the basic parameters in the attenuation fitting method can be given in advance based on the material's factory physicochemical parameters and conventional storage aging test experience.

[0047] Step S17: Substitute the continuous time-series monitoring dataset into the decay fitting method and calibrate the decay fitting method. For example, after importing all valid test sample data of the rudder component into the corresponding decay fitting method, the initial values ​​of the basic parameters are fine-tuned using the least squares iterative algorithm. The initial parameter calibration is completed when the overall residual converges to a preset accuracy threshold. In the verification stage, multiple sets of measured unfolding driving force data that did not participate in parameter training are randomly selected and compared with the fitted calculation values ​​at the corresponding time in the decay fitting method. Discrete data points with errors exceeding the threshold are traced and investigated to distinguish whether the data anomalies are caused by accidental experimental disturbances or insufficient model parameter adaptation. The parameter assignment is then iteratively corrected again to address the parameter adaptation defects.

[0048] Step S18: Based on the calibrated attenuation fitting method, output the dynamic time-series attenuation variable of the continuous torsion spring deployment driving force with storage time, and obtain the real-time attenuation value of the deployment driving force and the attenuation rate change law during the entire storage period.

[0049] In other words, the final attenuation fitting method after verification and optimization can output the attenuation curve of the deployment driving force under accelerated test conditions, and can also extrapolate and calculate the magnitude of the torsion spring deployment driving force and the instantaneous attenuation rate corresponding to any sampling node within the entire natural storage period based on the equivalent conversion relationship between accelerated aging and natural storage. This attenuation result serves as the prerequisite input condition for the subsequent deduction of the resistance degradation of the articulated pair and locking mechanism, and is used for the analysis and calculation in step S20.

[0050] It is worth noting that the grey model interpolation in step S14 is to expand sparse measured points into dense time-series data, while the decay fitting in steps S15 to S18 is to upgrade the mathematical prediction result into an engineering decay model that conforms to the physical aging law. The two have different purposes and belong to two-layer modeling (data layer → physical layer), rather than repeated fitting.

[0051] Step S20: Based on the degradation mechanism of each resistance element inside the rudder assembly, simulate the degradation law of each resistance element inside the rudder assembly with the storage period; wherein, the resistance elements inside the rudder assembly include the articulated joint and the locking mechanism, and the degradation mechanism of the resistance elements is the random fluctuation and nonlinear degradation characteristics of the resistance caused by the performance change of the articulated joint and the locking mechanism under the long-term storage environment.

[0052] This step can be further divided into steps S21 to S24 as follows.

[0053] Step S21: Retrieve the time-series environmental operating condition parameters within the preset storage period, construct the correlation model of storage environment operating condition parameters-structural degradation-resistance increment, and obtain the deterioration coefficient of the hinge pair and the deterioration coefficient of the locking mechanism based on the time-series environmental operating condition parameters.

[0054] Step S22: Substitute the deterioration coefficient of the articulated joint into the nonlinear quantitative relationship between the cumulative deterioration and the frictional resistance, and combine it with the sampling nodes that are synchronized with the attenuation time of the torsion spring's unfolding driving force to determine the basic increment of the frictional resistance of the articulated joint and the random fluctuation of the environment at each sampling node.

[0055] For example: Establish a cumulative degradation function: ;in, Storage time Cumulative deterioration of the articulated joint; , where is the limit of saturation degradation, and is a calibration constant; Storage time The deterioration coefficient of the hinged joint, It is a natural constant.

[0056] The basic increment of frictional resistance is calculated as follows, where the cumulative deterioration is linearly related to the frictional resistance, and a conversion factor is determined by experimental calibration: ;in, This is the basic increment of frictional resistance. This is the wear-resistance conversion factor.

[0057] Minor disturbances in the storage environment can cause random fluctuations in resistance. Gaussian random variables are used for simulation, and the fluctuation amplitude is adaptively adjusted according to the severity of the environment. ;in, Storage time The increase in total frictional resistance of the articulated joint; It represents random fluctuations in the environment, following a Gaussian distribution.

[0058] Step S23: Update the dynamic fluctuation range of the dynamic jamming resistance based on the deterioration coefficient of the locking mechanism. Combined with the irreversible deterioration mechanism of the locking mechanism, introduce a random mutation coefficient (following an exponential or uniform distribution) to simulate and generate the instantaneous abnormal peak value of the jamming resistance.

[0059] For example: the dynamic fluctuation range of dynamic hysteresis resistance is Then we have: ;in, Storage time The lower limit of the dynamic fluctuation range. Storage time The upper limit of the dynamic fluctuation range; This represents the lower limit of the initial resistance range, and is a fixed calibration value. This represents the upper limit of the initial resistance range and is a fixed calibration value. Storage time The deterioration coefficient of the locking mechanism.

[0060] The formula for calculating abnormal peak values ​​is: ;in, Storage time Abnormal peak value, This represents the random mutation coefficient.

[0061] Final storage time The total jamming resistance of the locking mechanism is: ;in, Storage time Total jamming resistance of the locking mechanism, when Dynamic fluctuation range exceeding dynamic hysteresis resistance If the mutation is triggered at a certain time, it is considered to have been triggered; otherwise, it is considered not to have been triggered.

[0062] Therefore, based on storage time Increase in total frictional resistance and storage time of articulated joint The total jamming resistance of the locking mechanism can be used to obtain the deterioration time series curves of the frictional resistance of the hinge pair and the jamming resistance of the locking mechanism.

[0063] Step S24: Integrate the basic increment of the frictional resistance of the articulated pair and the random fluctuation of the environment under each sampling node, as well as the dynamic fluctuation range and instantaneous abnormal peak value of the dynamic jamming resistance of the locking mechanism, to obtain the deterioration time sequence curve of the frictional resistance of the articulated pair and the deterioration time sequence curve of the jamming resistance of the locking mechanism, so as to realize the time-synchronous deduction of the structural resistance deterioration process and the deterioration process of the torsion spring deployment driving force.

[0064] Time-series environmental operating parameters include ambient temperature, ambient humidity, and storage time. (Equivalent storage years).

[0065] The correlation model is built based on measured data and takes environmental conditions as input. Its core function is to dynamically output two types of degradation coefficients according to real-time environmental conditions. The environmental condition parameters at each moment are input into the correlation model, and the model outputs the degradation coefficients of the articulated joint and the locking mechanism in real time according to the condition matching relationship calibrated by the experiment; the more severe the environmental conditions, the larger the corresponding coefficient value.

[0066] Step S30: Based on the mechanical transmission logic of the rudder assembly, the remaining driving force after the real-time torsion spring deployment driving force offsets the real-time articulated pair friction resistance and the real-time locking mechanism jamming resistance is used to simulate the actual effective deployment driving force of the rudder assembly.

[0067] This step can be further divided into steps S31 to S34 as follows.

[0068] Step S31: Organize the step-by-step mechanical transmission path of the rudder assembly to construct the mechanical priority logic of torsion spring deployment driving force, overcoming the frictional resistance of the articulated pair and overcoming the locking resistance to the effective driving force output.

[0069] The specific mechanical transmission logic is as follows: the deployment driving force generated by the release of the torsion spring's stored energy is the active driving force for the deployment of the rudder assembly, which is first used to overcome the sliding friction resistance of the articulated joint; after offsetting the friction resistance of the articulated joint, the remaining residual force is then used to overcome the jamming resistance of the locking mechanism, and finally the remaining driving force after completely offsetting the two levels of resistance is the effective deployment driving force that the rudder assembly can actually use.

[0070] Step S32: The real-time torsion spring deployment driving force, real-time hinge pair friction resistance, and real-time locking mechanism jamming resistance of the same sampling node are matched in time sequence.

[0071] The real-time torsion spring deployment driving force corresponding to different sampling nodes is taken from the real-time attenuation value of the deployment driving force throughout the entire storage cycle; the real-time articulated joint friction resistance and the real-time locking mechanism jamming resistance corresponding to each sampling node are taken from the degradation time-series curves of the articulated joint friction resistance and the locking mechanism jamming resistance obtained by the final fitting. This embodiment establishes a time-series dynamic mechanical cancellation calculation logic, aligning and matching the three heterogeneous parameters of driving force, friction resistance, and jamming resistance one by one according to the same storage reduction time node, ensuring that the driving capacity and resistance load corresponding to each calculation moment are the real degradation data under the same storage state.

[0072] Step S33: Based on the mechanical transmission path, calculate the mechanical difference step by step to offset the basic increment of the frictional resistance of the hinge pair and the random fluctuation of the environment, as well as the dynamic fluctuation range and the abnormal peak value of the instantaneous jamming resistance of the locking mechanism.

[0073] The step-by-step cancellation method reproduces the actual force process of the rudder assembly during deployment. It covers both the smooth nonlinear wear and deterioration loss of the articulated joint and the intermittent sudden loss caused by the irreversible deterioration of the locking mechanism. It characterizes all the resistance deterioration characteristics of the output and closely matches the actual storage deterioration conditions of the rudder assembly.

[0074] Step S34: Iteratively update the effective expansion driving force values ​​of each sampling node throughout the entire storage cycle in real time to form a continuous time-series change dataset of driving force deteriorating bidirectionally over time.

[0075] For example, iterative calculations can be performed on all sampling nodes throughout the entire storage period, continuously updating the final effective deployment driving force value at each moment. During long-term storage, the torsion spring deployment driving force continuously decays over time, while the structural resistance of the articulated joint and locking mechanism continuously increases over time, resulting in a two-way degradation effect of weakened driving capability and increased load resistance. Through full-time iterative calculations, a complete time-series dataset of effective driving force covering the entire storage period is finally constructed, intuitively representing the two-way degradation evolution of the rudder component's driving force with storage time, providing continuous and time-aligned core input data for subsequent degradation trajectory fitting and reliability evaluation.

[0076] Step S40: Combining the time-series decay variable of the torsion spring deployment driving force with the degradation law of the internal resistance element of the rudder assembly, extract the single-index degradation curves corresponding to the time-series sequences of mechanical margin, response speed and positioning reliability, and perform correlation fusion on each single-index degradation curve to form the comprehensive degradation trajectory of the rudder assembly deployment performance.

[0077] Step S40 can be further subdivided into at least the following steps S41 to S43.

[0078] Step S41: Extract the core degradation variables, which include the time-series decay variable of the torsion spring deployment driving force, the degradation law of the frictional resistance of the hinge pair with the storage period, and the degradation law of the jamming resistance of the locking mechanism with the storage period.

[0079] This step is used to select data aligned across the three time axes to form the data basis for subsequent coupling analysis.

[0080] Step S42: Extract the differential degradation features corresponding to the time series of mechanical margin, response speed and on-time reliability, and perform time series fitting to construct the corresponding single-index degradation curve.

[0081] Among them, the mechanical margin characterizes the power reserve capacity of the rudder assembly during deployment, the response speed characterizes the speed of deployment, and the positioning reliability characterizes whether the deployment action of the rudder assembly can be completed accurately. Specifically, the mechanical margin time series, response speed time series, and positioning reliability time series are generated by coupling the same set of drive degradation and drag degradation data, but they have completely different storage degradation characteristics: the mechanical margin is mainly affected by the continuous decay of the torsion spring deployment driving force and the steady-state increase of structural drag, showing a smooth degradation characteristic of continuous monotonicity and a gradually slowing decay rate; the response speed shows a steady lag degradation characteristic as the effective driving force weakens and the frictional drag accumulates and increases; the positioning reliability, in addition to being affected by steady-state degradation, is highly susceptible to the instantaneous jamming peak of the locking mechanism, showing a non-uniform degradation characteristic of steady degradation superimposed with local sudden fluctuations.

[0082] A: The timing fitting method for the mechanical margin is as follows: The mechanical margin is affected by the stress relaxation of the torsion spring and the steady-state accumulation of structural resistance, exhibiting a smooth deterioration pattern of monotonic decay and eventual saturation convergence. Mechanical margin = real-time deployment driving force of the torsion spring - frictional resistance of the hinge pair - jamming resistance of the locking mechanism. An exponential decay time-series fitting formula is used: In the formula, Storage time The real-time mechanical margin (i.e., the time margin of step S30) The actual effective deployment driving force); This represents the initial mechanical margin; The mechanical performance attenuation coefficient is jointly determined by the aging rate of the torsion spring and the steady-state growth rate of the structural resistance. This represents the steady-state compensation for the residual mechanical margin after long-term storage. The time-series fitting method for the mechanical margin characterizes the nonlinear smooth degradation characteristics of the mechanical margin, which is characterized by rapid deterioration in the early stage, a slower decay rate in the later stage, and a gradual stabilization. The mechanical margin is the difference between the effective driving force and the internal structural resistance. It is affected not only by the decay of the torsion spring but also by the dual influence of friction in the hinge pair and the increase in locking resistance. Therefore, its decay is faster, and it will eventually converge to a residual margin. .

[0083] B: The timing fitting method for response speed is as follows: The response speed exhibits a steady and continuous hysteresis degradation with increasing storage time, weakening driving force, and accumulation of frictional resistance, without sudden jumps. A quadratic polynomial time series fitting formula is used: In the formula, Storage time The response speed, i.e., the unfolding response time; This is the initial standard response rate, i.e., the standard response time; , The response degradation fitting coefficients are obtained through least-squares iteration using full-time sampled data. The time-series fitting method for response speed adapts to the degradation pattern of slow, continuous, and nonlinear lag in response speed, representing the steady-state degradation trajectory of gradually slowing action response under long-term storage.

[0084] C: The timing fitting method for on-time reliability is as follows: The reliability at the moment of arrival exhibits both long-term steady-state degradation and instantaneous jamming abrupt degradation characteristics. The reliability at each moment is determined by the base value of the effective driving force at that moment minus the abrupt deduction value determined by the peak jamming at that moment. Conventional single-fit formulas cannot simultaneously preserve both steady-state trends and abrupt peaks. This embodiment employs a two-layer coupled calculation formula combining steady-state basis fitting and abrupt threshold correction: ;in, Storage time The reliability of its arrival; For the steady-state base component, an exponential decay formula is used to fit the long-term slow degradation trend; For the instantaneous mutation correction component, when the locking resistance of the sampling node exceeds the normal steady-state fluctuation range, the instantaneous drop peak of reliability is obtained by substituting the random mutation coefficient. This term is 0 when the locking threshold is not triggered. The on-time reliability fitting method can completely retain two types of degradation characteristics: it fits the long-term stable degradation trend of reliability and accurately reproduces the instantaneous locking degradation mutation characteristics caused by environmental conditions, avoiding the erasure of real mutation failure information by conventional smooth fitting.

[0085] Using the three sets of differentiated calculation formulas mentioned above, the three types of performance time series are iteratively fitted point by point, resulting in three independent and mechanistically matched single-index degradation curves, providing curve data for subsequent multi-index fusion correction.

[0086] Step S43 involves correlating and fusing the degradation curves of individual indicators to correct degradation coupling deviations among multiple indicators, thus forming a unified comprehensive degradation trajectory for the rudder component's deployed performance. Specific implementation may include: Step S43-1: For the three single-index degradation curves , , To perform dimensionless normalization calculations, all indicators are mapped to the same numerical range. The general normalization formula is: Where m = 1, 2, 3; The value of the m-th indicator over the entire storage period. This represents the minimum value of the m-th indicator over the entire storage period; The value is the normalized index value.

[0087] Step S43-2: Correcting multi-index coupling deviation. The three performance indices of the rudder component exhibit mechanical coupling effects. Degradation of a single index can cause mutual interference and calculation deviations. A coupling correction coefficient is introduced to correct the deviations in the normalized results. In the formula: The coupling correction coefficient for the corresponding index is used to offset the degradation bias caused by the mutual influence between the indexes by accelerating the calibration of storage test data. This is the index value after deviation correction.

[0088] Step S43-3: Weighted fusion to generate comprehensive performance values Based on the importance of the three indicators to the deployment performance of the rudder assembly, weighting coefficients are configured. , and ; .

[0089] ; Step S43-4: Fit the integrated degradation trajectory, calculated for all sampling nodes in the full time series. The sequence is subjected to curve fitting to ultimately form the comprehensive degradation trajectory of the rudder component's deployment performance.

[0090] Furthermore, the method of this embodiment may further include: Step S50: By identifying critical decay nodes and abrupt deterioration nodes in the overall performance degradation trajectory, the complete evolution process of the rudder component performance from normal, weakened to critical failure is characterized.

[0091] In this step, based on the fitted comprehensive degradation trajectory of the rudder assembly, the performance evolution nodes throughout the entire storage cycle are identified and marked. Specifically, the moment when the rudder assembly's performance begins to show a sustained, slight weakening and deviates from its initial steady-state performance is marked as the critical decay node, characterizing the rudder assembly entering a slow performance degradation stage. The moment when the rudder assembly's performance drops sharply and fluctuates abruptly due to instantaneous jamming of the locking mechanism or sudden changes in resistance is marked as the abrupt degradation node, corresponding to occasional high-risk degradation conditions during storage. Through the complete marking of these two types of key nodes, the steady-state slow degradation process and the instantaneous abrupt degradation process of the rudder assembly's performance can be clearly distinguished, covering the entire lifecycle evolution of the rudder assembly from its initial normal state, intermediate performance weakening state, to its later critical failure state. This provides intuitive and accurate rudder assembly degradation data support for subsequent rudder assembly life prediction, failure mechanism localization, and reliability grading evaluation.

[0092] Furthermore, the method of this embodiment may also include evaluating the storage reliability of the rudder assembly through the following steps: Step S60: Pre-store the standard failure thresholds for the mechanical margin, response speed, and positioning reliability of the rudder assembly, and then match the performance degradation trajectory of the rudder assembly with the failure thresholds.

[0093] This step involves comparing the measured and extrapolated values ​​of mechanical margin, response speed, and on-time reliability at each sampling node throughout the entire storage cycle in the fitted rudder component's comprehensive performance degradation trajectory with the pre-stored corresponding failure thresholds point by point. This marks the degree of deviation of each performance indicator from the failure threshold in the current storage stage, clarifies the current state range of each performance of the rudder component, and establishes a judgment benchmark for subsequent lifespan extrapolation and risk assessment.

[0094] Among them, the mechanical margin failure threshold is the minimum power reserve value required for the rudder assembly to complete the deployment action normally. When the real-time mechanical margin is lower than this value, it is determined that the power supply is insufficient. The response speed failure threshold is the maximum allowable deployment time of the rudder assembly. When the actual deployment time exceeds this value, it is determined that the action response lag exceeds the standard. The arrival reliability failure threshold is the lower limit of the percentage of qualified action completion. When the reliability index is lower than this threshold, it is determined that the deployment action cannot be completed stably and accurately.

[0095] Step S70: The time point at which the performance of the rudder component reaches the failure threshold is estimated by using the trajectory extrapolation method, so as to predict the remaining reliable storage life of the rudder component.

[0096] Using existing full-cycle comprehensive degradation trajectories as basic data, segments within the trajectories exhibiting a stable trend of performance degradation are selected as extrapolation sample intervals. Based on the time-series patterns of continuous changes in performance indicators, a time-series trend extrapolation method is employed to extend and extrapolate the degradation trajectory, continuously calculating the mechanical margin, response speed, and on-time reliability values ​​corresponding to different future storage durations.

[0097] During extrapolation, the extrapolated performance values ​​are compared with preset failure thresholds in real time. When the extrapolation result first reaches the failure threshold of any performance indicator, the total storage time at that moment is recorded. The remaining reliable storage life of the rudder component is obtained by subtracting the actual storage time from this total time. If multiple indicators reach the failure threshold successively, the indicator that reaches the failure threshold first is used as the judgment criterion to ensure that the life prediction result fits the actual failure logic. At the same time, the extrapolation results are further validated for reasonableness, and temporary out-of-bounds data caused by local abrupt deterioration is removed. Only the steady-state deterioration trend is used as the core basis for life prediction, thereby improving the accuracy of the prediction.

[0098] Optionally, the calculation formula corresponding to the trajectory extrapolation method is: ;in, Let k be the virtual storage duration for the kth phase of the extrapolation stage, where k = 1, 2, 3, ...; Indicate duration The corresponding extrapolated performance index is calculated as one of mechanical margin, response speed and on-time reliability. The single-index degradation function obtained by fitting each extrapolated performance index separately (i.e., the one mentioned above) , and ).

[0099] The formula for calculating the remaining reliable storage life of the rudder assembly is: ;in, For the remaining reliable storage life of the rudder assembly, The total storage duration corresponding to the first time the extrapolated performance index reaches the standard failure threshold. This refers to the actual storage time of the rudder component.

[0100] Among these, jamming of the locking mechanism can cause a sudden and abrupt drop in the index. Such temporary fluctuations do not represent an overall steady-state failure. Therefore, fluctuation verification conditions should be set to filter out invalid abrupt changes. The filtering conditions for fluctuation verification can be as follows: ;in, Duration The moving average of the performance metrics corresponding to each stage; A preset threshold for normal fluctuations is established. When the index value exceeds the range of the above inequality, it is determined to be an instantaneous abrupt deterioration point and is not included in the steady-state failure determination. Only when multiple consecutive extrapolated sampling points stably meet the corresponding determination conditions (for example, multiple consecutive extrapolated sampling points of mechanical margin stably meet the standard failure threshold, multiple consecutive extrapolated sampling points of response speed stably meet the standard failure threshold, and multiple consecutive extrapolated sampling points of on-time reliability stably meet the standard failure threshold), is it formally determined to have reached a failure state, and the corresponding limit storage time is recorded.

[0101] During extrapolation, the extrapolated performance values ​​are compared with the preset failure threshold in real time. Combined with mutation filtering rules, temporary out-of-bounds data caused by local mutation deterioration are eliminated. Only the steady-state deterioration trend is used as the core basis for lifetime extrapolation to improve prediction accuracy.

[0102] Step S80: Quantify and calculate the contribution of torsion spring preload degradation and internal drag element degradation to the performance degradation of the rudder assembly.

[0103] Based on the fitted performance time-series degradation curves, the performance degradation values ​​corresponding to the decrease in torsion spring driving force and the increase in structural resistance were statistically analyzed. The two types of degradation values ​​were then proportionally converted to the total overall performance degradation value to obtain the corresponding percentage of degradation contribution. This percentage ranges from 0% to 100%; a higher percentage indicates a stronger driving effect of this factor on the overall performance degradation.

[0104] Step S90: Based on the proportion of degradation contribution, distinguish between core failure causes and secondary influencing factors, and complete the storage reliability classification evaluation and failure risk location of the rudder component.

[0105] Among the deterioration of torsion springs and resistance components, the one with a higher proportion is the core failure cause, and the one with a lower proportion is the secondary influencing factor; when the difference between the two proportions is within a preset range, it is determined that the two types of factors jointly dominate the deterioration.

[0106] For example, based on the degradation contribution ratio obtained in the previous step, the causative factors are classified into levels: degradation factors with a higher contribution ratio are identified as core failure causative factors, while those with a lower contribution ratio are identified as secondary influencing factors. Simultaneously, the overall storage reliability of the rudder assembly is graded based on the degree of deviation between performance indicators and failure thresholds, and the remaining reliable storage life.

[0107] Generally, it can be divided into three levels: Level 1 is a high reliability state, where all performance indicators are far above the failure threshold, the remaining storage life is sufficient, and the impact of both types of degradation factors is at a low level; Level 2 is a performance weakening state, where some indicators are gradually approaching the failure threshold, the remaining life is moderate, and the core degradation causes need to be addressed specifically; Level 3 is a critical risk state, where performance indicators are close to the failure threshold, the remaining reliable life is short, the core failure causes have had a significant negative impact, and there is a risk of short-term failure.

[0108] While completing the classification, risk identification is performed based on the core failure causes: if the contribution of torsion spring preload deterioration is high, the root cause of the risk is determined to be stress relaxation and aging of the torsion spring material; if the contribution of internal resistance element deterioration is high, the root cause of the risk is determined to be cracking of the articulated joint structure and deterioration of the locking mechanism structure. Based on the identification results, a complete reliability evaluation conclusion and risk warning are output, providing clear data support and decision-making basis for rudder component storage management, batch screening, and structural optimization.

[0109] It is worth noting that the analysis results of steps S80 and S90 above can be used to detect and determine the above configuration weight coefficients. , and Whether it is scientifically sound and reasonable, if the deviation reaches the preset target, the above-mentioned configuration weight coefficients can be readjusted based on the deviation value. , and As a variation, when configuring the weighting coefficients in step S43, the degradation contribution ratio result in step S80 is referenced in advance to make a more reasonable initial assignment.

[0110] In summary, the method in this embodiment effectively solves the technical problems of existing methods for analyzing the storage performance degradation of rudder components, such as the difficulty in implementing disassembly and actual testing, the difficulty in implementing multi-component coupled analysis, and the problems of distortion of performance degradation analysis results, low accuracy of life prediction, and poor reliability when using simulation methods. A detailed analysis follows: Firstly, this embodiment relies on real storage time-series monitoring data of the torsion spring, a core component affecting the deployment function of the rudder assembly. Combined with the degradation mechanism of the hinged joints and locking mechanisms of the various resistance elements within the rudder assembly, it simulates the degradation pattern of each resistance element within the rudder assembly over the storage period. Then, based on the mechanical transmission logic of the rudder assembly, it simulates the actual effective deployment driving force of the rudder assembly. Simultaneously, combining the time-series decay variable of the torsion spring deployment driving force with the degradation pattern of the internal resistance elements of the rudder assembly, it extracts the single-index degradation curves corresponding to the time-series sequences of mechanical margin, response speed, and positioning reliability. These single-index degradation curves are then correlated and fused to form a comprehensive degradation trajectory of the rudder assembly's deployment performance. Furthermore, by identifying critical decay nodes and abrupt degradation nodes in the comprehensive degradation trajectory, it characterizes the complete evolution process of the rudder assembly's performance from normal, weakened, to critical failure. Moreover, it can combine the single-index degradation curves to complete the storage reliability evaluation of the rudder assembly.

[0111] Therefore, this embodiment is based on the timing monitoring data of the torsion spring deployment driving force that can be collected in situ. The entire process does not require disassembly of the rudder assembly. It can complete the non-destructive storage degradation analysis without damaging the product structure and storage state, effectively reducing the testing cost. It is suitable for engineering scenarios of long-term storage reliability assessment of batch products, and has stronger practicality and scalability.

[0112] Secondly, this embodiment uses measured data on the degradation of the driving force as the core driving force, and combines it with the actual storage degradation mechanism of the internal resistance elements of the rudder assembly to simultaneously deduce the collaborative degradation law of the articulated joint and the locking mechanism, thus reproducing the unsteady degradation characteristics of multiple resistance elements. This effectively makes up for the technical defect of the traditional simulation results being out of touch with the actual working conditions.

[0113] Furthermore, this embodiment overcomes the limitations of single-factor degradation analysis that only analyzes the torsion spring of the rudder assembly, and achieves a two-factor coupled and collaborative analysis of torsion spring preload degradation and structural drag degradation.

[0114] Finally, this embodiment constructs a comprehensive performance degradation trajectory coupled with two deterioration factors, which can clearly describe the entire evolution of the rudder component from normal operation and performance degradation to critical failure.

[0115] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for analyzing the performance degradation of a rudder assembly, characterized in that, Includes the following steps: Step S10: Obtain the timing monitoring data of the torsion spring deployment driving force of the rudder assembly within the preset storage period. Based on the timing monitoring data of the torsion spring deployment driving force, fit the timing decay variable of the torsion spring deployment driving force with the storage period. Step S20: Based on the degradation mechanism of each resistance element inside the rudder assembly, simulate the degradation law of each resistance element inside the rudder assembly with the storage period; wherein, the resistance element inside the rudder assembly includes the articulated joint and the locking mechanism, and the degradation mechanism of the resistance element is the random fluctuation and nonlinear degradation characteristics of the resistance caused by the performance change of the articulated joint and the locking mechanism under the long-term storage environment. Step S30: Based on the mechanical transmission logic of the rudder assembly, the remaining driving force after the real-time torsion spring deployment driving force offsets the real-time articulated pair friction resistance and the real-time locking mechanism jamming resistance is used to simulate the actual effective deployment driving force of the rudder assembly. Step S40: Combining the time-series decay variable of the torsion spring deployment driving force with the degradation law of the internal resistance element of the rudder assembly, extract the single-index degradation curves corresponding to the time-series sequences of mechanical margin, response speed and positioning reliability, and perform correlation fusion on each single-index degradation curve to form the comprehensive degradation trajectory of the rudder assembly deployment performance.

2. The method for analyzing the performance degradation of a rudder assembly according to claim 1, characterized in that, Step S10 includes: Step S11: Determine the complete storage cycle based on the preset storage life of the rudder component and set the sampling nodes for time-series monitoring; Step S12: At each sampling node, the in-situ multiple rounds of retesting and collection of the deployment driving force corresponding to the built-in torsion spring of the rudder assembly are performed to obtain multiple sets of original mechanical test data. Step S13: Preprocess the original mechanical test data, remove failed and distorted data by combining the preset normal performance range of the rudder component, and complete the data noise reduction optimization by selecting the best of multiple sets of retest data. Step S14: Correlate the preprocessed effective deployment driving force data with the corresponding storage reduction time in a time series to construct a continuous time series monitoring dataset that forms a time-mechanical performance correspondence. Step S15: Extract the completed continuous time-series monitoring dataset, and build a basic data sample set for time-series fitting with storage time as the independent variable and the preprocessed torsion spring unfolding driving force as the dependent variable. Step S16: Determine the attenuation fitting method by combining the stress relaxation and constitutive characteristics of the torsion spring metal material during long-term storage and aging, and determine the initial values ​​of the basic parameters of the attenuation fitting method. Step S17: Substitute the continuous time-series monitoring dataset into the attenuation fitting method and calibrate the attenuation fitting method. Step S18: Based on the calibrated attenuation fitting method, output the dynamic time-series attenuation variable of the continuous torsion spring deployment driving force with storage time, and obtain the real-time attenuation value of the deployment driving force and the attenuation rate change law during the entire storage period.

3. The method for analyzing the performance degradation of a rudder assembly according to claim 1, characterized in that, Step S20 includes: Step S21: Retrieve the time-series environmental operating condition parameters within the preset storage period, construct the correlation model of storage environment operating condition parameters-structural degradation-resistance increment, and obtain the deterioration coefficient of the hinge pair and the deterioration coefficient of the locking mechanism based on the time-series environmental operating condition parameters. Step S22: Substitute the deterioration coefficient of the articulated joint into the nonlinear quantitative relationship between the cumulative deterioration and the frictional resistance, and combine it with the sampling nodes that are synchronized with the attenuation time of the torsion spring's unfolding driving force to determine the basic increment of the frictional resistance of the articulated joint and the amount of environmental random fluctuation at each sampling node. Step S23: Update the dynamic fluctuation range of the dynamic jamming resistance based on the deterioration coefficient of the locking mechanism. Combined with the irreversible deterioration mechanism of the locking mechanism, introduce a random mutation coefficient to simulate and generate the instantaneous abnormal peak value of the jamming resistance. Step S24: Integrate the basic increment of the frictional resistance of the articulated pair and the random fluctuation of the environment under each sampling node, as well as the dynamic fluctuation range and instantaneous abnormal peak value of the dynamic jamming resistance of the locking mechanism, to obtain the deterioration time sequence curve of the frictional resistance of the articulated pair and the deterioration time sequence curve of the jamming resistance of the locking mechanism, so as to realize the time-series synchronous deduction of the structural resistance deterioration process and the torsion spring deployment driving force deterioration process.

4. The method for analyzing the performance degradation of a rudder assembly according to claim 1, characterized in that, Step S30 includes: Step S31: Organize the step-by-step mechanical transmission path of the rudder assembly to construct the mechanical priority logic of torsion spring deployment driving force, overcoming the frictional resistance of the articulated pair and overcoming the locking resistance to the effective driving force output. Step S32: Time-match the real-time torsion spring deployment driving force, real-time hinge pair friction resistance, and real-time locking mechanism jamming resistance of the same sampling node. Step S33: According to the mechanical transmission path, perform mechanical difference calculation step by step to offset the basic increment of frictional resistance of the articulated pair and the random fluctuation of the environment, as well as the dynamic fluctuation range and instantaneous abnormal peak value of dynamic jamming resistance of the locking mechanism. Step S34: Iteratively update the effective expansion driving force values ​​of each sampling node throughout the entire storage cycle in real time to form a continuous time-series change dataset of driving force deteriorating bidirectionally over time.

5. The method for analyzing the performance degradation of a rudder assembly according to claim 1, characterized in that, Step S40 includes: Step S41: Extract the core degradation variables, including the time-series decay variable of the torsion spring deployment driving force, the degradation law of the frictional resistance of the hinge pair with the storage period, and the degradation law of the jamming resistance of the locking mechanism with the storage period. Step S42: Extract the differential degradation features corresponding to the time series of mechanical margin, response speed and on-time reliability, and perform time series fitting to construct the corresponding single-index degradation curve; Step S43: Perform correlation fusion on the single-index degradation curves, correct the degradation coupling deviation between multiple indices, and form a unified comprehensive degradation trajectory of the rudder component's deployment performance.

6. The method for analyzing the performance degradation of a rudder assembly according to claim 5, characterized in that, Step S40 also includes: Step S44: By identifying critical decay nodes and abrupt deterioration nodes in the integrated degradation trajectory, the complete evolution process of the rudder component performance from normal, weakened to critical failure is characterized.

7. The method for analyzing the performance degradation of a rudder assembly according to claim 5, characterized in that, Also includes: Step S60: Pre-store the standard failure thresholds for the mechanical margin, response speed, and positioning reliability of the rudder assembly, and then match the performance degradation trajectory of the rudder assembly with the failure thresholds. Step S70: Use the trajectory extrapolation method to estimate the time node when the performance of the rudder component reaches the failure threshold, so as to predict the remaining reliable storage life of the rudder component.

8. The method for analyzing the performance degradation of a rudder assembly according to any one of claims 1 to 7, characterized in that, Also includes: Step S80: Quantify and calculate the contribution of torsion spring preload degradation and internal drag element degradation to the performance degradation of the rudder assembly, respectively. Step S90: Based on the proportion of degradation contribution, distinguish between core failure causes and secondary influencing factors, and complete the storage reliability classification evaluation and failure risk location of the rudder component.

Images