Maintenance-oriented flight control hydraulic actuator protection ring reliability evaluation method

Abstract

The invention provides a maintenance-oriented flight control hydraulic actuator protection ring reliability evaluation method, the method is applied to a hydraulic actuator protection ring, and the method comprises the following steps: constructing an evaluation method for different flight time protection ring performance indexes; screening a protection ring test piece, and performing sample manufacturing and performance testing; carrying out calculation analysis on the test data by utilizing a single-factor ANOVA multiple comparison algorithm or a multi-sample median equality algorithm; and fitting the performance degradation curve of the protection ring, establishing a performance degradation model, and predicting the reliability of the protection ring. According to the method, the reliability of the protection ring can be effectively evaluated, and design or process personnel can make decisions on parts which need to be replaced for the protection ring in the stages of rebuilding, prolonging the service life and the like of the hydraulic actuator according to the analysis result, so that the maintenance cost is saved and the maintenance period is shortened on the premise of ensuring the reliability of the product.

Description

Technical Field

[0001] The invention belongs to the research field of reliability maintenance technology of flight control hydraulic actuators, and specifically relates to a maintenance-oriented flight control hydraulic actuator protection ring reliability evaluation method. Background Art

[0002] The hydraulic actuator is an electro-hydraulic servo mechanism that uses aviation hydraulic oil as the working medium and controls the deflection of the aircraft's control surface according to the instructions issued by the flight control computer. The protective ring is also called a "protective ring" or a "retaining ring". Its structural forms are divided into two types: non-cut and cut. The protective ring is usually used in conjunction with an O-type rubber seal ring to form a sealing structure for the hydraulic actuator. Its function is to prevent the O-type rubber seal ring from being squeezed into the matching gap between the sealing cylinder and the shaft, and to reduce the installation deformation and wear of the rubber seal ring. The materials of the protective ring are usually polytetrafluoroethylene sheets, rods, heat recovery sheets, etc.

[0003] In recent years, in response to the general requirement of high-quality development of the aviation industry, research on low-cost and high-reliability maintenance technology has become one of the important research directions. There are hundreds of various types of protective rings installed on hydraulic actuators. During the overhaul and life extension of actuator products, there is currently no clear basis or conclusion on whether the protective rings need to be replaced as a must-have part.

[0004] At present, most of the performance degradation studies on actuator protection rings in China are material-level aging tests, and no reliability studies have been conducted on the finished protection rings after the actuators are used (including installation and life tests). However, the complex working conditions on aircraft, such as working medium, high temperature and high pressure, extreme temperature shock, alternating loads, etc., are quite different from the aging test conditions of protection rings. Therefore, it is necessary to study a method to evaluate the reliability of protection rings after use, establish a performance degradation model, and predict the reliability of protection rings. Summary of the invention

[0005] The purpose of the present invention is to study a maintenance-oriented flight control hydraulic actuator protection ring reliability evaluation method to solve the problem of whether the protection ring needs to be a necessary replacement part in different stages such as hydraulic actuator product overhaul and life extension, and whether the protection ring cannot be evaluated or the evaluation is inaccurate.

[0006] The technical solution of the present invention is a maintenance-oriented flight control hydraulic actuator protection ring reliability evaluation method, comprising the following steps:

[0007] Step 1: Determine the research object, research method and performance indicators. The research object is the flight control hydraulic actuator protection ring. The research method adopts hypothesis testing in statistics (single-factor ANOVA multiple comparison algorithm or multi-sample median equality algorithm) and regression analysis. The performance indicators of the research are the main performance indicators specified in the protection ring material specification - tensile force and elongation at break.

[0008] Step 2: Screening of protection ring samples and sample quantity. The protection ring samples to be studied must meet the requirements of being installed with the hydraulic actuator or undergoing life tests with the hydraulic actuator, and the continuous working time must be between the first overhaul (first overhaul) and the second overhaul (second overhaul) of the hydraulic actuator.

[0009] A group of unused new protection ring samples are selected as the reference samples, and the protection rings that have been installed or have undergone life tests with hydraulic actuators are selected as the test sample group.

[0010] The number of samples in each group shall be calculated based on the minimum sample size required for hypothesis testing; when the calculated minimum sample size exceeds the actual number of installations of the protection ring on each hydraulic actuator, at least 30% of the number of installations of the protection ring shall be used as the minimum sample size for each group of samples.

[0011] Step 3: Preparation before the test. Remove the protective ring from the actuator. For the protective ring with a cut, it can be directly disassembled; for the protective ring without a cut, cut along the cutting line between the inner and outer surfaces of the protective ring, and the angle between the cutting line and the diameter of the inner or outer surface is 45°. After disassembly, wipe the protective ring with absorbent cotton dipped in anhydrous alcohol, and number the protective rings of each group of samples one by one.

[0012] Step 4: Test. Install the protective ring on the tensile test bench, set the initial tensile gauge length, stretch at a constant speed along both ends of the cut of the protective ring until the protective ring breaks, and measure and record the tensile force and elongation at break when the protective ring breaks.

[0013] Step 5: Independence test. The independence of the tensile force and elongation at break measurement data of each group of protective ring samples is tested by the median run test method. The run test method arranges the measurement data of each group of samples, takes the median as the boundary, and records those exceeding the median as "+", and those less than or equal to the median as "-". The part with continuous symbols is called a run, and then the total number of measurement data runs r is calculated.

[0014] When the total number of runs r satisfies the "lower critical value R1<r<upper critical value R2" (R1, R2 are obtained by looking up the total number of runs test table), it indicates that the measured data are independent of each other. When the test data of each group of samples are independent of each other, proceed to step 6. If any group of data in tensile force and elongation at break is not independent, it is necessary to analyze the measured data, eliminate abnormal values, and add the number of samples in the corresponding sample group to re-test.

[0015] Outliers are the largest number of consecutive “+” data that exceed the median or the largest number of consecutive “-” data that are lower than the median in a run. For example, in a set of measurement data with a total run number r=5 (“++---+++++--++”), when the total run number r=5 does not meet the upper and lower critical values, the “+++++” data are outliers.

[0016] Step 6: Normality test. The premise of the single-factor ANOVA multiple comparison algorithm is that the measured data conforms to the normal distribution, so it is necessary to perform a normality test on the measured data of each group of samples. The test method uses the Anderson-Dar ling cumulative distribution function ECDF algorithm for calculation, which is the most effective of all normality test methods currently available. When the test result P1 value is greater than 0.05, it indicates that the measured data conforms to the normal distribution, and proceed to step 7; otherwise, proceed to step 8.

[0017] Step 7: Single-factor ANOVA multiple comparison calculation and analysis. Single-factor ANOVA multiple comparison calculation and analysis is a method for hypothesis testing and judging whether there are statistical differences in the mean values ​​of measurement data of multiple sample groups, and the overall error rate is controlled at 5%. When the calculation result P2 value is less than 0.05, it indicates that there are differences between the mean values ​​of measurement data of different sample groups; then the Tukey method can be used to group the mean values ​​of measurement data of each group of samples. The grouping results show that there are significant differences between sample groups without the same letters, and there are no significant differences between sample groups with the same letters.

[0018] When there is no significant difference in the tensile force or elongation at break of each group of samples, it is considered that the tensile force or elongation at break performance of the protection ring has not deteriorated with the extension of the use time, and the performance of the protection ring is reliable.

[0019] When there is a significant difference in tensile force or elongation at break between the reference sample group and the test sample group, but there is no significant difference in tensile force or elongation at break between the test sample groups, it is considered that with the extension of the use time, the tensile force or elongation at break performance has declined but the decline is not obvious, and the performance reliability of the protective ring has declined but meets the use requirements.

[0020] There are significant differences in tensile force or elongation at break between the test sample groups. It is believed that with the extension of usage time, the tensile force or elongation at break performance decreases significantly, the performance reliability of the protective ring does not meet the usage requirements, and a new protective ring needs to be replaced.

[0021] Step 8: Calculation and analysis of the multi-sample median equality test algorithm. This method is a method for hypothesis testing and judging whether the medians of the measurement data of multiple sample groups are statistically different. It is suitable for calculation and analysis when the measurement data of the samples do not follow the normal distribution. When the calculation result P3 value is less than 0.05, it indicates that there are significant differences between different sample groups.

[0022] When there are significant differences or no significant differences in the tensile force and elongation at break between the groups of samples, between the reference sample group and the test sample group, and between the test sample groups, the evaluation method for the reliability of the protection ring is the same as step seven.

[0023] Step 9: Regression analysis. Establish the performance degradation fitting curve and curve regression equation of the mean or median of the tensile force and elongation at break of each group of protection ring samples and the flight time, test the significance and validity of the curve regression equation, and predict the reliability of the protection ring performance. When the significance test P4 value of the curve regression equation is less than 0.05, the fitting curve is valid, that is, the performance of the protection ring decreases with the increase of flight time; the total effect R of the regression equation 2 The larger the value, the more ideal the fitting curve.

[0024] Beneficial effects of the invention: The technical solution of the invention proposes a new research idea and method for the reliability evaluation of the protection ring of the flight control hydraulic actuator, establishes a reliability evaluation method and technical indicators, and solves the problem that whether the protection ring needs to be a necessary replacement part cannot be evaluated or the evaluation is inaccurate at different stages such as overhaul and life extension of the hydraulic actuator.

[0025] Advantages of the present invention:

[0026] 1. A maintenance-oriented reliability evaluation method for the protection ring of the flight control hydraulic actuator was constructed, and the screening requirements, production methods, and test methods for the test samples for the reliability research of the protection ring finished product were proposed.

[0027] 2. Reliability evaluation indicators for the finished product of the flight control hydraulic actuator protection ring are proposed, including the tensile strength and elongation at break of the protection ring at different flight times.

[0028] 3. A guard ring performance degradation fitting curve was established to achieve reliability prediction of the guard ring performance.

[0029] 4. The method of the present invention can provide a reference idea for the research on reliability maintenance technology of other components of flight control hydraulic actuators. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 is a flow chart of the method of the present invention;

[0031] Figure 2 This is the protection ring pattern of the flight control hydraulic actuator. DETAILED DESCRIPTION

[0032] In order to make the technical solution and effects of the present invention more clearly described, Figure 1 And implementation cases are provided to provide a detailed and complete description of the technical solutions in the present invention.

[0033] Embodiment 1

[0034] Step 1: Determine the research object, analysis method and performance index. The research object is the XX-22-57 protection ring after installation and use of a certain type of hydraulic actuator. The research method adopts hypothesis testing in statistics (single-factor ANOVA multiple comparison algorithm or multi-sample median equality test algorithm) and regression analysis method. The performance indicators of the research are tensile strength and elongation at break.

[0035] Step 2: Screening of protection ring samples and sample quantity. The selected XX-22-57 protection ring samples with continuous flight time of installation are 0h, 1188h, 1727h, 1821h, and 2190h, and the number is 10 for each (the actual number of installations does not reach the minimum sample size required for hypothesis testing calculation, so 62.5% of the actual number of installations of each actuator is sampled), of which the protection ring sample of 0h is a new protection ring, which is used as the benchmark sample group, and the protection rings of other flight times are used as the test sample group.

[0036] Step 3: Preparation before the test. Remove the protective ring from the actuator parts, use special tools to cut the XX-22-57 protective ring, and cut along the cutting line between the inner and outer surfaces of the protective ring. The angle between the cutting line and the diameter of the inner or outer surface is 45°. After disassembly, wipe the protective ring with absorbent cotton dipped in anhydrous alcohol, and number the protective rings of each group of samples from "1 to 10".

[0037] Step 4: Test. Install the protective ring on a tensile test bench with an accuracy of ±0.5%, set the initial tensile gauge length to 20 mm, stretch at a constant speed along both ends of the cut of the protective ring until the protective ring breaks, and measure and record the tensile force and elongation at break when the protective ring breaks.

[0038] Step 5: Independence test. The independence of the tensile force and elongation at break measurement data of each group of protective ring samples was tested using the median run test method. The test results showed that the number of runs r of the median of each group of samples satisfied R1=2<r<R2=10, and the measurement data were independent of each other.

[0039] Step 6: Normality test, the test method uses Anderson-Dar ling cumulative distribution function ECDF algorithm for calculation. The P1 value of the test results of the tensile force and elongation at break measurement data of each group of samples is > 0.05, and the measurement data conforms to the normal distribution.

[0040] Step 7: Single-factor ANOVA multiple comparison calculation and analysis. Taking the average tensile force F and elongation at break of each group of samples as the response, the flight time t as the level, and the overall error rate controlled at 5%, the average value of the measurement data of each group of samples (see Table 1) was subjected to single-factor ANOVA multiple comparison calculation and analysis. The calculation results P2 values ​​were all <0.05, indicating that there were differences in the average values ​​of the measurement data of each group of five different flight time protection rings compared with each other.

[0041] Tukey's method was used to group the mean values ​​of the sample measurement data of each group. Sample groups without the same letters indicated significant differences, and sample groups with the same letters had no significant differences (see Table 2).

[0042] Table 1XX-22-57 Average values ​​of sample measurement data for each group

[0043]

[0044] Table 2 Tukey method grouping

[0045]

[0046] According to the grouping results in Table 2, there is a significant difference in tensile force between the XX-22-57 protective ring benchmark sample group and the test sample group, but there is no significant difference in tensile force between the test sample groups. It is believed that with the extension of use time, the tensile force performance of the protective ring has decreased but the decrease is not obvious, and the reliability of the tensile force performance has decreased but meets the use requirements.

[0047] There is no significant difference in the elongation at break of each group of samples of the XX-22-57 protection ring (although the protection ring of 1821h is different from other sample groups in terms of grouping results, the average elongation at break is greater than that of the new protection ring as shown in Table 1, which meets the technical specifications of the protection ring. Therefore, it can be considered that there is no difference from the new protection ring). It is believed that with the extension of the use time, the elongation at break performance of the protection ring will not decrease, and the elongation at break performance of the protection ring is reliable.

[0048] Step 9: Regression analysis and significance test. Establish the performance degradation fitting curve of the average value of the tensile force and elongation at break measurement data of each group of protection ring samples and the flight time, and test the significance and validity of the curve regression equation.

[0049] The regression equation of the fitting curve of the tensile force performance degradation of the protective ring is: F = 54.90-0.008232t+0.000002t 2 , regression equation significance test P4 value = 0.005 < 0.05, regression equation total effect R 2=99.5%, the curve fitting is significantly effective. It can be seen from the curve that with the increase of flight time, the tensile force performance of the protection ring decreases.

[0050] The regression equation of the elongation at break performance degradation fitting curve is: δ = 183.0 + 0.00419t, the regression equation significance test P4 value = 0.926 > 0.05, and the total effect of the regression equation R 2 =0.3%, the significance and effectiveness of the curve fitting are very small, and the curve fitting is approximately a straight line, that is, with the increase of flight time, the elongation at break performance does not decrease significantly.

[0051] In summary, based on the performance comparison analysis and degradation model of each group of protection ring samples of XX-22-57, the performance of the protection ring of XX-22-57 is reliable before 2190 hours of installation and use, and no replacement is required.

[0052] Embodiment 2

[0053] Step 1: Determine the research object, analysis method and performance index. The research object is the XX-04-12 protection ring after installation and use of a certain type of hydraulic actuator. The research method adopts hypothesis testing in statistics (single-factor ANOVA multiple comparison algorithm or multi-sample median equality test algorithm) and regression analysis method. The performance indicators of the research are tensile strength and elongation at break.

[0054] Step 2: Screening protection ring samples and sample quantity. The selected XX-04-12 protection ring samples have installation durations of 0h, 1188h, 1727h, 1821h, and 2190h, with 10 samples each (the actual installation quantity does not reach the minimum sample quantity required for hypothesis testing calculation, so 33.3% of the actual installation quantity of each actuator is sampled), of which the protection ring sample of 0h is a new protection ring, which is used as the benchmark sample group, and the protection rings of other flight times are used as the test sample group.

[0055] Step 3: Preparation before the test. Remove the protective ring from the actuator parts, use special tools to cut the XX-04-12 protective ring, and cut along the cutting line between the inner and outer surfaces of the protective ring. The angle between the cutting line and the diameter of the inner or outer surface is 45°. After disassembly, wipe the protective ring with absorbent cotton dipped in anhydrous alcohol, and number the protective rings of each group of samples one by one from "1 to 10".

[0056] Step 4: Test. Install the protective ring on a tensile test bench with an accuracy of ±0.5%, set the initial tensile gauge length to 20 mm, stretch at a constant speed along both ends of the cut of the protective ring until the protective ring breaks, and measure and record the tensile force and elongation at break when the protective ring breaks.

[0057] Step 5: Independence test, the independence of the tensile force and elongation at break measurement data of each group of protective ring samples is tested, and the test method adopts the median run test method. The number of runs r of the median of the test results all satisfies R1=2<r<R2=10, and the measurement data are independent of each other.

[0058] Step 6: Normality test, the test method uses Anderson-Dar ling cumulative distribution function ECDF algorithm for calculation. Test results 1821h protection ring elongation at break data P1 value <0.05, the measured data does not conform to the normal distribution.

[0059] Step 8: Calculation and analysis of the multi-sample median equality test algorithm. Taking the median of the tensile force F and elongation at break δ measurement data of each group of samples (see Table 3) as the response and the flight time t as the level, the multi-sample median equality test algorithm calculation and analysis are performed. The analysis results are as follows:

[0060] Table 3XX-04-12 Median of sample measurement data for each group

[0061]

[0062] The P3 value of the multi-sample median equality test of the tensile force of the benchmark sample group and the test sample group is less than 0.05, indicating that there are differences in the median of the measured data of the tensile force of the five different flight time protection rings. However, the P3 value of the tensile force between the test sample groups is greater than 0.05, indicating that there is no significant difference in the tensile force between the test sample groups. It is believed that with the extension of the use time, the tensile force performance of the protection ring has decreased but the decrease is not obvious, and the reliability of the tensile force performance has decreased but meets the use requirements.

[0063] The P3 value of the elongation at break of the reference sample group and the test sample group is greater than 0.05, indicating that there is no difference in the median of the elongation at break data of each group. It is believed that with the extension of the use time, the elongation at break performance of the protection ring does not decrease, and the elongation at break performance of the protection ring is reliable.

[0064] Step 9: Regression analysis and significance test. Establish the performance degradation fitting curves of the tensile force, median elongation at break and flight time of each group of retaining ring samples, and test the significance and validity of the curve regression equation.

[0065] The regression equation of the tensile force performance degradation fitting curve of the XX-04-12 protective ring is: F = 45.27 + 0.005071t - 0.000011t 2 +0.000000t 3 , the significance test of regression equation P4 value = 0.007 < 0.05, the total effect of regression equation R 2=99.9%, the curve fitting is significantly effective. It can be seen from the curve that with the increase of flight time, the tensile force performance of the protection ring decreases.

[0066] The regression equation of the elongation at break performance degradation fitting curve is: δ = 104.7 + 0.001369t, the regression equation significance test P4 value = 0.804 > 0.05, and the regression equation total effect R 2 =2.4%, the significance and effectiveness of the curve fitting are low, and the curve fitting is approximately a straight line, that is, with the increase of flight time, the elongation at break performance does not decrease significantly.

[0067] In summary, based on the performance comparison analysis and degradation model of each group of protection ring samples of XX-04-12, the performance of the XX-04-12 protection ring is reliable before 2190 hours of installation and use, and no replacement is required.

Claims

1. A maintenance-oriented flight control hydraulic actuator protection ring reliability evaluation method, characterized by: Here are the steps: Step 1: Screen the protection ring samples and determine the number of samples for each group; Step 2: Remove the protective ring from the actuator; Step 3: Perform a tensile test on the protective ring to measure the tensile force and elongation at break; Step 4: Conduct an independence test on the measurement data of each group of samples. If the measurement data of each group of samples are independent of each other, proceed to step 5. Otherwise, remove the outliers and add the number of samples in the corresponding sample group to retest. Step 5: Perform normality test on the measurement data of each group of samples respectively; if the measurement data of each group of samples conforms to normal distribution, proceed to step 6, otherwise proceed to step 7; Step 6: Perform one-way ANOVA multiple comparison analysis on the average values ​​of the measured data of each group of samples; determine whether there are significant differences between different sample groups, and determine the reliability of the protection ring based on whether there are significant differences; Step 7: Perform multi-sample median equality calculation and analysis on the median of the measured data of each group of samples; determine whether there are significant differences between different sample groups, and determine the reliability of the protection ring based on whether there are significant differences; Step 8: Perform regression analysis and significance test on the mean or median of the measurement data of each group of samples.

2. The method according to claim 1, characterized in that: In step 1, the samples include a reference sample group and a test sample group; the reference sample group selects unused new protection rings, and the test sample group selects protection rings whose installation time or life test time with the hydraulic actuator is between the first turn and the second turn; The sample size for each sample group is calculated according to the minimum sample size required for hypothesis testing.

3. The method according to claim 2, characterized in that: In step 2, the protective ring with a cut is directly disassembled, and the protective ring without a cut is cut along the cutting line between the inner and outer surfaces of the protective ring, and the angle between the cutting line and the diameter of the inner or outer surface intersection point is 45°.

4. The method according to claim 3, characterized in that: In step three, the protective ring is installed on the tensile test bench, the initial tensile gauge length is set, and the two ends of the cut of the protective ring are stretched until the protective ring breaks.

5. The method according to claim 4, characterized in that: In step 4, when the measurement data of each group of samples are within the range of upper and lower critical values, it indicates that the measurement data are independent of each other; when any group of measurement data in each group of samples is not independent, it is necessary to analyze the measurement data, eliminate outliers, and supplement the number of samples in the corresponding sample group to re-test.

6. The method according to claim 5, characterized in that: In step five, the Anderson-Darling cumulative distribution function ECDF algorithm is used to perform a normality test on the measurement data of each group of samples; when the test result P1>0.05, the measurement data conforms to the normal distribution.

7. The method according to claim 6, characterized in that: In step six, with tensile force and elongation at break as responses and flight time as level, the overall error rate was controlled at 5%, and the average values ​​of the measured data of each group of samples were subjected to single-factor ANOVA multiple comparison analysis. When the calculated result P2 < 0.05, it indicated that there were differences between different sample groups. Then, the Tukey algorithm was used to group the average values ​​of the measured data of each group of samples. There were significant differences between sample groups without the same letters, and there were no significant differences between sample groups with the same letters.

8. The method according to claim 7, characterized in that: In step seven, the median of each group of sample measurement data is subjected to multi-sample median equality calculation and analysis. When the calculated result P3 value is less than 0.05, it indicates that there are significant differences between the groups of samples.

9. The method according to claim 8, characterized in that: In step 6 and step 7, when there is no significant difference in the tensile force or elongation at break of each group of samples, it is considered that the tensile force or elongation at break performance of the protection ring has not degraded with the extension of the use time, and the performance of the protection ring is reliable; When there is a significant difference in tensile force or elongation at break between the reference sample group and the test sample group, but there is no significant difference in tensile force or elongation at break between the test sample groups, it is considered that with the extension of the use time, the tensile force or elongation at break performance has decreased but the decrease is not obvious, and the performance reliability of the protection ring has decreased but meets the use requirements; There are significant differences in tensile force or elongation at break between the test sample groups. It is believed that with the extension of usage time, the tensile force or elongation at break performance decreases significantly, the performance reliability of the protective ring does not meet the usage requirements, and a new protective ring needs to be replaced.

10. The method according to claim 9, characterized in that: In step eight, performance degradation fitting curves between the tensile force, the average or median of the elongation at break and the flight time are established respectively; and the significance and validity of the regression equation are tested respectively. When the significance test P4 value is less than 0.05, the fitting curve is valid, that is, as the flight time increases, the performance of the guard ring decreases. The total effect R 2 The larger the value, the more ideal the fitting curve.

Images