A method and apparatus for aircraft reliability assessment based on Gaussian model

Abstract

The present application belongs to the technical field of aircraft reliability design, and particularly relates to an aircraft reliability evaluation method and device based on a Gaussian model. The method comprises: step S1, obtaining the effective failure counts in a plurality of consecutive set time periods; step S2, forming a plurality of cumulative time periods in a manner that includes all previous set time periods, and determining the effective failure counts in each cumulative time period; step S3, determining the MTBF confidence lower limit in each cumulative time period; step S4, when the MTBF observation value in each cumulative time period is greater than the MTBF confidence lower limit, solving the coefficient of the Gaussian model; step S5, based on the Gaussian model with the determined coefficient, calculating the effective failure counts in the total time period including the next time period to be predicted; step S6, determining the effective failure counts in the next time period to be predicted. The present application uses the evaluated numerical values ​​to predict the failure conditions in future time periods, thereby realizing the continuous evaluation of the aircraft's mean trouble-free working time.

Description

Technical Field

[0001] The present application belongs to the technical field of aircraft reliability design, and in particular relates to an aircraft reliability evaluation method and device based on a Gaussian model. Background Art

[0002] In recent years, aviation equipment technology has developed vigorously. As a typical complex equipment, the reliability of aircraft is one of the important indicators for evaluating the quality of aircraft. Among them, the mean time between failures (MTBF) refers to the average working time between two adjacent failures, which is an extremely important evaluation indicator. The gradual application of new technologies and new products has led to an increasing requirement for aircraft reliability, so aircraft reliability assessment is crucial.

[0003] At present, there are many methods and research on reliability assessment in China, but in the direction of aircraft design, more emphasis is placed on the evaluation and analysis of reliability indicators of products and equipment, and less research is done on the reliability assessment of the entire aircraft. A large amount of historical data is generated during the daily use and maintenance of aircraft. Research on data-driven reliability assessment methods for entire aircraft is of great significance and military value.

[0004] The Chinese invention patent with publication number CN110675047A discloses an aircraft reliability assessment method based on assembly outline and fault list data, and calculates the mean time between failures (MTBF) through test flight data and process assembly outline. The MTBF index of the aircraft can be effectively calculated using statistical methods; the Chinese invention patent with publication number CN114065386A discloses a multivariate linear regression equation method for quickly predicting the MTBF of the whole machine. Based on the aircraft parameter data related to the engine, the MTBF calculation of the aircraft engine can be realized by establishing a multivariate linear regression model. The above methods can realize the calculation of MTBF, but they do not have the ability to perform trend analysis and integrated evaluation of the MTBF of the aircraft. Summary of the invention

[0005] In order to solve the above problems, the present application provides an aircraft reliability assessment method and device based on a Gaussian model, establishes a trend analysis model according to Gaussian theory, and uses the accumulation method to calculate the model parameters, thereby realizing the trend analysis of the MTBF of the whole aircraft and realizing subsequent fault prediction.

[0006] The first aspect of the present application provides an aircraft reliability assessment method based on a Gaussian model, which mainly includes:

[0007] Step S1, obtaining the number of effective faults counted within a plurality of consecutive set time periods;

[0008] Step S2, forming a plurality of cumulative time periods in a manner that includes all previously set time periods, and determining the number of valid faults in each cumulative time period;

[0009] Step S3, determining the MTBF confidence lower limit in each cumulative time period, where MTBF is the mean time between failures;

[0010] Step S4: When the MTBF observation value in each cumulative time period is greater than the MTBF lower confidence limit, the coefficient of the Gaussian model is solved to determine the Gaussian model coefficient;

[0011] Step S5, based on the Gaussian model of the determination coefficient, calculate the number of effective faults in the total time period including the next time period to be predicted and the previous multiple consecutive set time periods;

[0012] Step S6: Determine the number of valid faults in the next time period to be predicted.

[0013] Preferably, in step S2, the accumulated time periods include at least 3.

[0014] Preferably, step S3 further comprises:

[0015] Step S31, determining the MTBF observation value based on the number of effective failures counted within a cumulative time period;

[0016] Step S32, determining the lower confidence limit coefficient of the MTBF during reception;

[0017] Step S33: Determine the MTBF lower confidence limit based on the MTBF observation value and the MTBF lower confidence limit coefficient.

[0018] Preferably, in step S4, the Gaussian model is:

[0019] N i =b0+T i b1+(T i ) 2 b2;

[0020] Among them, N i =lnM i , M i For the i The number of effective faults in a cumulative time period, T i For the i The time of the cumulative time period, b0, b1, b2 are the Gaussian model coefficients.

[0021] The second aspect of the present application provides an aircraft reliability assessment device based on a Gaussian model, which mainly includes:

[0022] A module for obtaining the number of valid faults in a set time period, which is used to obtain the number of valid faults counted in multiple consecutive set time periods;

[0023] A cumulative time period valid fault count statistics module is used to form multiple cumulative time periods in a manner that includes all previously set time periods, and determine the valid fault count of each cumulative time period;

[0024] MTBF confidence lower limit calculation module, used to determine the MTBF confidence lower limit in each cumulative time period, where MTBF is the mean trouble-free working time;

[0025] The Gaussian model coefficient solving module is used to solve the coefficient of the Gaussian model and determine the Gaussian model coefficient when the MTBF observation value in each cumulative time period is greater than the MTBF confidence lower limit;

[0026] A module for calculating the number of effective faults in a total time period, which is used to calculate the number of effective faults in a total time period including the next time period to be predicted and a plurality of consecutive previous set time periods based on a Gaussian model with a determination coefficient;

[0027] The effective fault number calculation module for the time period to be predicted is used to determine the effective fault number for the next time period to be predicted.

[0028] Preferably, the accumulation time period includes at least 3.

[0029] Preferably, the MTBF confidence lower limit calculation module includes:

[0030] An MTBF observation value calculation unit, used to determine an MTBF observation value based on the number of effective failures counted within a cumulative time period;

[0031] A confidence lower limit coefficient calculation unit, used to determine the confidence lower limit coefficient of the MTBF during reception;

[0032] The confidence lower limit calculation unit is used to determine the MTBF confidence lower limit based on the MTBF observation value and the MTBF confidence lower limit coefficient.

[0033] Preferably, the Gaussian model is:

[0034] N i =b0+T i b1+(T i ) 2 b2;

[0035] Among them, N i =lnM i , M i For the i The number of effective faults in a cumulative time period, T i For thei The time of the cumulative time period, b0, b1, b2 are the Gaussian model coefficients.

[0036] This application uses the evaluated values ​​to predict fault conditions in future time periods and realizes continuous evaluation, which effectively solves the problem of aircraft MTBF indicator evaluation under long-term data accumulation during the actual use of equipment. It is easy and fast to use and has good engineering applicability and universality. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a flowchart of a preferred embodiment of the aircraft reliability assessment method based on the Gaussian model of the present application. Implementation

[0038] In order to make the purpose, technical scheme and advantages of the implementation of this application clearer, the technical scheme in the implementation of this application will be described in more detail in combination with the drawings in the implementation of this application. In the drawings, the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The described implementation is a part of the implementation of this application, not all of the implementations. The implementation described below with reference to the drawings is exemplary and is intended to be used to explain this application, and cannot be understood as a limitation on this application. Based on the implementation in this application, all other implementations obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. The implementation of this application is described in detail below in combination with the drawings.

[0039] This application provides an aircraft reliability assessment method based on a Gaussian model, such as Figure 1 As shown, it mainly includes:

[0040] Step S1, obtaining the number of effective faults counted within a plurality of consecutive set time periods;

[0041] Step S2, forming a plurality of cumulative time periods in a manner that includes all previously set time periods, and determining the number of valid faults in each cumulative time period;

[0042] Step S3, determining the MTBF confidence lower limit in each cumulative time period, where MTBF is the mean time between failures;

[0043] Step S4: When the MTBF observation value in each cumulative time period is greater than the MTBF lower confidence limit, the coefficient of the Gaussian model is solved to determine the Gaussian model coefficient;

[0044] Step S5, based on the Gaussian model of the determination coefficient, calculate the number of effective faults in the total time period including the next time period to be predicted and the previous multiple consecutive set time periods;

[0045] Step S6: Determine the number of valid faults in the next time period to be predicted.

[0046] In step S1 of the present application, Excel is first used to filter the aircraft quality control data to obtain valid data. Assuming that the calculation unit is day and the given time period is month, the calculation process of the implementation case is described based on the three-month hypothetical data. For example, the aircraft quality control data in the first month gave 4 valid faults, the second month gave 5 valid faults, and the third month gave 6 valid faults. When i=1, t=30, r=4; when i=2, t=30, r=5; when i=3, t=30, r=6. i is the month, r is the number of valid faults, and t is the set time period in days, that is, the set time period is 30 days.

[0047] In some optional implementations, in step S2, the cumulative time period includes at least 3.

[0048] In the above embodiment with specific parameters, the first month is accumulated separately, at this time, the cumulative time period T1=30, and the number of valid faults M1=4; the first month and the second month are accumulated separately, at this time, the cumulative time period T2=60, and the number of valid faults M2=9; the three months are accumulated together, at this time, the cumulative time period T3=90, and the number of valid faults M3=15.

[0049] In step S3, the MTBF confidence lower limit in each cumulative time period is determined respectively. In some optional implementations, step S3 further includes:

[0050] Step S31, determining the MTBF observation value based on the number of effective failures counted within a cumulative time period;

[0051] Step S32, determining the lower confidence limit coefficient of the MTBF during reception;

[0052] Step S33: Determine the MTBF lower confidence limit based on the MTBF observation value and the MTBF lower confidence limit coefficient.

[0053] In this step, the observed value (point estimate) of MTBF is calculated using formula (1): , and then use formula (2) to calculate the MTBF confidence lower limit coefficient during reception .

[0054] (1);

[0055] (2);

[0056] Among them, T is the total working time of the object, r is the number of failures of the object, and C is the confidence level, which is given by humans. is x with degree of freedom i2 The upper gamma quantile of the distribution can be determined by looking up a table of related digits.

[0057] Finally, use formula (3) to calculate the MTBF confidence limit .

[0058] ;

[0059] For the effective failure times counted in each cumulative time period in step S2, for the first cumulative time period, calculate the observed value (point estimate) of MTBF according to formula (1): is 7.5; according to formula (2), C' is calculated to be 0.025, and the MTBF confidence lower limit coefficient is 0.391; use formula (3) to calculate the MTBF confidence lower limit For the second accumulation period, calculate the observed value (point estimate) of MTBF The MTBF confidence limit is 6.7. For the third cumulative period, calculate the observed value (point estimate) of MTBF: is 6, MTBF lower confidence limit It is 3.638.

[0060] Since in the above calculation results, the MTBF observation values ​​in each cumulative time period are all greater than the MTBF confidence lower limit, indicating that the data is credible, the Gaussian model coefficients can be determined through step S4.

[0061] In some optional implementations, in step S4, the Gaussian model is:

[0062] N i =b0+T i b1+(T i ) 2 b2;

[0063] Among them, N i =lnM i , M i For the i The number of effective faults in a cumulative time period, T i For the i The time of the cumulative time period, b0, b1, b2 are the Gaussian model coefficients.

[0064] Substituting the values ​​in step S2, we can solve: b0 is 0.28, b1 is 0.042, and b2 is -0.00017.

[0065] The Gaussian model in step S4 is described below. The Gaussian model used in this step is actually a deformation of the existing Gaussian model. The existing Gaussian model is:

[0066] (4).

[0067] Among them, M max is the peak value of Gaussian fitting, T max is the peak position of Gaussian fitting, and S is the half-width information.

[0068] Set N i =lnM i ,b0=lnM max -(T max ) 2 / S,b1=2T max / S, b2=-1 / S, then we can get the following formula.

[0069] N i =b0+T i b1+(T i ) 2 b2(5).

[0070] The above formula is the deformation formula of the Gaussian model. The model coefficients can be solved in step S4 through the above formula, and then used in step S5. Of course, if the original Gaussian model needs to be used, the above formula can be recorded in the form of N=XB and solved according to the least squares principle. The characteristic parameters of the following original Gaussian model can be obtained.

[0071] (6).

[0072] Based on the Gaussian model or the transformed formula, in step S5, data such as the number of effective faults predicted to occur in the fourth cumulative time period can be obtained as follows:

[0073] When i=4, T4=120, and the calculated value of M4 is 18. In step S6, the effective failure times of the first three months are subtracted to obtain the effective failure times of the fourth month r4=3.

[0074] In addition to the one month given in the above embodiment, the setting time period of step S1 of this application can also be one week, half a month, one quarter, one year, etc., which can be set according to actual conditions. This application is based on 2 years, uses aircraft quality monitoring and maintenance data for calculation, and performs trend analysis on the cumulative MTBF and its lower confidence limit. It is verified that the data fits well. The aircraft MTBF evaluation and analysis method based on the Gaussian model constructed by this application is effective and has good stability. This application effectively solves the problem of aircraft MTBF indicator evaluation under long-term data accumulation during the actual use of equipment. It is easy and fast to use and has good engineering applicability and universality.

[0075] The second aspect of the present application provides an aircraft reliability assessment device based on a Gaussian model corresponding to the above method, mainly comprising:

[0076] A module for obtaining the number of valid faults in a set time period, which is used to obtain the number of valid faults counted in multiple consecutive set time periods;

[0077] A cumulative time period valid fault count statistics module is used to form multiple cumulative time periods in a manner that includes all previously set time periods, and determine the valid fault count of each cumulative time period;

[0078] MTBF confidence lower limit calculation module, used to determine the MTBF confidence lower limit in each cumulative time period, where MTBF is the mean trouble-free working time;

[0079] The Gaussian model coefficient solving module is used to solve the coefficient of the Gaussian model and determine the Gaussian model coefficient when the MTBF observation value in each cumulative time period is greater than the MTBF confidence lower limit;

[0080] A module for calculating the number of effective faults in a total time period, which is used to calculate the number of effective faults in a total time period including the next time period to be predicted and a plurality of consecutive previous set time periods based on a Gaussian model with a determination coefficient;

[0081] The effective fault number calculation module for the time period to be predicted is used to determine the effective fault number for the next time period to be predicted.

[0082] In some optional implementations, the cumulative time period includes at least 3.

[0083] In some optional implementations, the MTBF confidence lower limit calculation module includes:

[0084] An MTBF observation value calculation unit, used to determine an MTBF observation value based on the number of effective failures counted within a cumulative time period;

[0085] A confidence lower limit coefficient calculation unit, used to determine the confidence lower limit coefficient of the MTBF during reception;

[0086] The confidence lower limit calculation unit is used to determine the MTBF confidence lower limit based on the MTBF observation value and the MTBF confidence lower limit coefficient.

[0087] In some optional implementations, the Gaussian model is:

[0088] N i =b0+T i b1+(T i ) 2 b2;

[0089] Among them, N i =lnM i , M i For the i The number of effective faults in a cumulative time period, T i For the i The time of the cumulative time period, b0, b1, b2 are the Gaussian model coefficients.

[0090] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. An aircraft reliability assessment method based on a Gaussian model, characterized in that: include: Step S1, obtaining the number of effective faults counted within a plurality of consecutive set time periods; Step S2, forming a plurality of cumulative time periods in a manner that includes all previously set time periods, and determining the number of valid faults in each cumulative time period, wherein the cumulative time periods include at least three; Step S3, determining the MTBF confidence lower limit in each cumulative time period, where MTBF is the mean time between failures; Step S4: When the MTBF observation value in each cumulative time period is greater than the MTBF lower confidence limit, the coefficient of the Gaussian model is solved to determine the Gaussian model coefficient; Step S5, based on the Gaussian model of the determination coefficient, calculate the number of effective faults in the total time period including the next time period to be predicted and the previous multiple consecutive set time periods; Step S6, determining the number of valid faults in the next time period to be predicted; Wherein, in step S4, the Gaussian model is: Among them, N i =lnM i , M i is the number of effective failures in the ith cumulative time period, T i is the time of the ith cumulative time period, b0, b1, b2 are the Gaussian model coefficients.

2. The aircraft reliability assessment method based on Gaussian model as claimed in claim 1, characterized in that: Step S3 further comprises: Step S31, determining the MTBF observation value based on the number of effective failures counted within a cumulative time period; Step S32, determining the lower confidence limit coefficient of the MTBF during reception; Step S33: Determine the MTBF lower confidence limit based on the MTBF observation value and the MTBF lower confidence limit coefficient.

3. An aircraft reliability assessment device based on a Gaussian model, characterized in that: include: A module for obtaining the number of valid faults in a set time period, which is used to obtain the number of valid faults counted in multiple consecutive set time periods; A cumulative time period valid fault count statistics module, used to form multiple cumulative time periods in a manner that includes all previously set time periods, and determine the valid fault count of each cumulative time period, wherein the cumulative time periods include at least 3; MTBF confidence lower limit calculation module, used to determine the MTBF confidence lower limit in each cumulative time period, where MTBF is the mean trouble-free working time; The Gaussian model coefficient solving module is used to solve the coefficient of the Gaussian model and determine the Gaussian model coefficient when the MTBF observation value in each cumulative time period is greater than the MTBF confidence lower limit; A module for calculating the number of effective faults in a total time period, which is used to calculate the number of effective faults in a total time period including the next time period to be predicted and a plurality of consecutive previous set time periods based on a Gaussian model with a determination coefficient; A valid fault count calculation module for a time period to be predicted, used to determine the valid fault count for the next time period to be predicted; Wherein, the Gaussian model is: Among them, N i =lnM i , M i is the number of effective failures in the ith cumulative time period, T i is the time of the ith cumulative time period, b0, b1, b2 are the Gaussian model coefficients.

4. The aircraft reliability assessment device based on Gaussian model as claimed in claim 3, characterized in that: The MTBF confidence lower limit calculation module includes: An MTBF observation value calculation unit, used to determine an MTBF observation value based on the number of effective failures counted within a cumulative time period; A confidence lower limit coefficient calculation unit, used to determine the confidence lower limit coefficient of the MTBF during reception; The confidence lower limit calculation unit is used to determine the MTBF confidence lower limit based on the MTBF observation value and the MTBF confidence lower limit coefficient.

Images