System testability prediction method based on aero-engine functional architecture

Abstract

A system testability prediction method based on an aero-engine functional architecture comprises the following steps: determining system test architecture data according to the aero-engine functional architecture; based on the system test architecture data, obtaining a fault mode probability lambda Di capable of being correctly tested by the ith basic-level maintainable function unit and a fault rate lambda i of the ith basic-level maintainable function unit, and establishing a system fault detection rate rFD mathematical model; obtaining a function fault mode probability lambda Ii which can be isolated to an ith basic level maintainable function unit, and establishing a system fault isolation rate rFI mathematical model; obtaining a fault mode probability lambda Ai that the i-th basic-level maintainable function unit with the test function may generate a false alarm, and establishing a system inherent false alarm rate rFA mathematical model; and system testability prediction is completed. The method can be used for scheme design of aero-engines and other complex products, system testability prediction in the detailed design stage and evaluation of whether the scheme design of the aero-engines meets testability quantitative index requirements or not.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of general quality characteristics of an aero-engine, and particularly relates to a system testability prediction method based on a functional architecture of an aero-engine. BACKGROUND

[0002] The testability level of an aero-engine is generally measured by a fault detection rate, a fault isolation rate and a false alarm rate, and the testability level of the aero-engine directly affects the rapid repair capability of an aircraft. With the development of electronic control and test technology, the aero-engine increasingly uses online automatic test means, so that the testability level of the aero-engine is continuously improved.

[0003] At present, the existing national standards and national military standards lack modeling and prediction methods of testability based on a functional architecture. In related technical documents, only the topmost theoretical model is disclosed, and the relationship between the functional architecture, the system test functional architecture and the testability index is not disclosed. The testability modeling methods and cases that can be applied in engineering internationally are mainly based on a multi-signal flow model. On the one hand, the model needs to be established on the basis of a determined "fault-test" signal flow relationship, and the corresponding model can be established only after the detailed design of the test system is completed, and the signal flow modeling analysis is performed by means of software; on the other hand, these models are mainly applicable to the testability design at the circuit level, and it is assumed that the test end / point is fault-free, and the testability design at multiple levels of a complex mechanical and electrical functional architecture cannot be described. Engineering practice shows that all functional units of the test functional architecture have a certain fault rate. The fault detection rate and the fault isolation rate calculated by the model are too ideal and deviate from the engineering practice, and it is difficult to apply the model in a complex mechanical and electrical integrated product such as an aero-engine. This leads to the fact that the testability prediction and evaluation based on the existing theoretical methods cannot be carried out in the early stage of the scheme design and engineering development of the aero-engine, and the functional architecture design cannot be ensured to meet the quantitative index requirements of testability, thereby greatly increasing the risk of subsequent testability design.

[0004] Therefore, there is an urgent need to design a technical scheme that can be applied to the testability analysis and evaluation of the architecture design results in the early stage of the scheme design or detailed design of an aero-engine, evaluate whether the scheme design can meet the quantitative index requirements of testability, and disclose the relationship between the functional architecture, the system test functional architecture and the testability index. SUMMARY

[0005] To solve the above technical problems, the application provides a system testability prediction method based on a functional architecture of an aero-engine, which comprises the following steps:

[0006] According to the functional architecture of the aero-engine, system test architecture data is determined, and the system test architecture data comprises the functional type of a product level object, an automatic testability decision value D Oij , an independent test decision value DDij And the self-test capability value B of the testing function ij The functional types include control execution functions, testing functions, and data acquisition and self-testing functions.

[0007] Based on the system test architecture data, obtain the probability of the fault mode that the i-th basic-level repairable functional unit can be correctly tested and the failure rate λ of the i-th basic-level repairable functional unit. i Establish system fault detection rate r FD Mathematical model;

[0008] Based on the obtained system test architecture data and the probability λ of the functional failure modes that can be correctly tested in the basic-level repairable functional units, Di Obtain the functional failure mode probability λ that can be isolated down to the i-th basic level maintainable functional unit. Ii Establish system fault isolation rate r FI Mathematical model;

[0009] Based on the obtained system test architecture data and the probability λ of the functional failure modes that can be correctly tested in the basic-level repairable functional units, Di Obtain the probability λ of the failure mode that may produce a false alarm for the i-th basic-level repairable functional unit with testing function. Ai Establish the system's inherent false alarm rate r FA Mathematical model;

[0010] Based on the established system fault detection rate r FD Mathematical model, system fault isolation rate r FI Mathematical model, inherent false alarm rate r of the system FA A mathematical model is used to complete the system testability prediction, which includes the system failure detection rate r. FD Expected system fault isolation rate r FI Expected and inherent false alarm rate r FA Expected.

[0011] Furthermore, the system fault detection rate r FD The mathematical model is as follows:

[0012]

[0013] Where i = 1, 2, 3, ..., n, and n is the number of basic-level maintainable functional units.

[0014] Furthermore, the probability λ of the fault modes that the i-th basic-level maintainable functional unit can be correctly tested is... Di The mathematical model is as follows:

[0015]

[0016] wherein λ Dij — the probability of correct testable fault mode of the jth control execution class function in the ith base level maintainable function unit, j = 1, 2, 3, …, Q Di , Q Di is the number of control execution class functions in the ith base level maintainable function unit;

[0017] λ Tik — the probability of correct testable fault mode of the kth test class function in the ith base level maintainable function unit, k = 1, 2, 3, …, Q Ti , Q Ti is the number of test class functions in the ith base level maintainable function unit;

[0018] λ Bil — the probability of correct testable fault mode of the lth collection and self-test function in the ith base level maintainable function unit, l = 1, 2, 3, …, Q Bi , Q Bi is the number of collection and self-test functions in the ith base level maintainable function unit.

[0019] Further, the probability of correct testable fault mode of the jth control execution class function in the ith base level maintainable function unit λ Dij comprises:

[0020] the probability of correct testable fault mode of the jth control execution class function in the ith base level maintainable function unit λ Dij ,

[0021] the probability of correct testable fault mode of the kth test class function in the ith base level maintainable function unit λ Dij ,

[0022] the probability of correct testable fault mode of the lth collection and self-test function in the ith base level maintainable function unit λ Dij ,

[0023] wherein D Oij — the automatic testability decision value of the jth control execution class function in the ith base level maintainable function unit; α Oij — the proportion of fault modes of the jth control execution class function in the ith base level maintainable function unit; α Dik — the proportion of the kth test function fault mode corresponding to the jth control execution class function in the ith base level maintainable function unit; λKBi — the failure rate of the kth test function in the ith base-level maintainable functional unit in the acquisition and self-test functional unit set by the superior system; a KBiq — the proportion of the failure mode of the corresponding acquisition and self-test function in the acquisition and self-test functional unit set by the superior system for the kth test function in the ith base-level maintainable functional unit; R k — the redundancy of the corresponding kth test function of the jth control and execution class function in the ith base-level maintainable functional unit; t s — the working time of the base-level maintainable functional unit in the average mission flight time; a Sip — the proportion of the pth test function failure mode of the corresponding system-level test unit of the jth control and execution class function in the ith base-level maintainable functional unit; l Si — the failure rate of the testable function in the corresponding functional test unit set by the superior system in the ith base-level maintainable functional unit; l SBi — the failure rate of the pth test function of the corresponding system-level test unit of the jth control and execution class function in the ith base-level maintainable functional unit in the acquisition and self-test functional unit set by the superior system; a SBiq — the proportion of the failure mode of the corresponding acquisition and self-test function in the acquisition and self-test functional unit set by the superior system for the pth test function of the corresponding system-level test unit of the jth control and execution class function in the ith base-level maintainable functional unit; R p — the redundancy of the pth test function of the corresponding Si system-level test unit of the jth control and execution class function in the ith base-level maintainable functional unit;

[0024] Further, the system failure isolation rate r FI The mathematical model is:

[0025]

[0026] Further, the functional failure mode probability l Ii comprises:

[0027] The failure mode probability l IDij of the jth control and execution class function in the ith base-level maintainable functional unit is isolated by setting a test function by itself

[0028] l IDij = D Dij l Dij ;

[0029] The fault mode probability λISij of the jth control execution class function in the ith base-level maintainable function unit can be isolated by setting the test function unit at the system level,

[0030]

[0031] The fault mode probability λ of the kth test class function in the ith base-level maintainable function unit can be isolated by setting the self-test function unit at the system level, ITik ,

[0032]

[0033] When the base-level maintainable function unit contains system acquisition and self-test functions, the fault mode probability λ of the lth acquisition and self-test function in the ith base-level maintainable function unit can be isolated, IBil ,

[0034]

[0035] wherein nr is the number of maintainable function units associated with the jth control execution class function in the ith base-level maintainable function unit in a function-related system, nr≤n; r is the number of control execution class functions associated with the jth control execution class function in the ith base-level maintainable function unit in the maintainable function units associated with the function-related system, r≤QDi; λ Bi is the failure rate of the acquisition and self-test function of the maintainable function unit set in the upper-level system corresponding to the kth test function in the ith base-level maintainable function unit; α Biq is the proportion of the fault mode of the acquisition and self-test function in the maintainable function unit set in the upper-level system corresponding to the kth test function in the ith base-level maintainable function unit, q=1, 2, 3, …, Q Bi , Q Bi is the number of acquisition and self-test class functions in the Bi th system-level maintainable function unit.

[0036] Further, the system inherent false alarm rate r FA The mathematical model is:

[0037]

[0038] Further, the fault mode probability λ of the ith base-level maintainable function unit with test function that may generate false alarm is Ai The mathematical model is:

[0039]

[0040] wherein λ AikThe false alarm failure mode probability of the kth test type function of the ith base-level maintainable function unit with test function is λ Ti , k = 1, 2, 3,..., Q Ti , Q Aik is the number of test type functions in the ith base-level maintainable function unit.

[0041] Further, when the base-level maintainable function unit has test function, the false alarm failure mode probability λ Aik of the kth test type function of the ith base-level maintainable function unit with test function is:

[0042] λ Dik = α i F BAik ;

[0043] wherein F BAik is the self-test failure probability of the kth test function of the ith base-level maintainable function unit, mainly considering the self-test failure probability caused by the self-test function failure of the part with self-test capability.

[0044] Further, the self-test failure probability F BAik of the kth test function of the ith base-level maintainable function unit is:

[0045]

[0046] The present application has the beneficial effects that: through the proposed system testability prediction method based on the function architecture of an aero-engine, the data relationship between the multi-level test architecture, function testability, function failure mode probability and fault detection, isolation and false alarm of the base-level maintainable function unit is revealed based on the function architecture of the product level of the corresponding system and function unit of the aero-engine, the mathematical model of the system fault detection rate, the system fault isolation rate and the system inherent false alarm rate is constructed, the prediction of the system fault detection rate r FD , the system fault isolation rate r FI and the system inherent false alarm rate r FA is performed, the system testability prediction is completed, and the testability modeling and prediction in the scheme design and detailed design stages of the aero-engine and other complex products can be used, the scheme design of the aero-engine can be evaluated whether it can meet the testability quantitative index requirement, and the relationship between the function architecture, the system test function architecture and the testability index can be revealed. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 is the flowchart of the system testability prediction method based on the function architecture of an aero-engine provided by the present application;

[0048] Figure 2is a typical system test function architecture model block diagram provided by the present application. DETAILED DESCRIPTION

[0049] The technical solutions of the present application are further described below, but the scope of protection is not limited to the description.

[0050] The embodiment of the present application provides a system testability prediction method based on an aero-engine function architecture, as shown in the figure, comprising the following steps: Figure 1

[0051] Step S100, according to the aero-engine function architecture, system test architecture data is determined, the system test architecture data comprises the function type of the product level object, the automatic testability decision value D Oij , the independent test decision value D Dij and the test function self-test capability value B ij ; the function type comprises a control execution class function, a test class function and a collection and self-test class function.

[0052] The system test function architecture of the aero-engine scheme design stage is the basis for establishing the testability model and performing the testability prediction. The system test function architecture is based on the function architecture, and is based on the test architecture related data such as the basic level maintainable function unit and the function transmission relationship. A typical system test function architecture model block diagram is shown in the figure. Figure 2

[0053] The following data needs to be determined on the basis of the system test function architecture model:

[0054] a, the function type of the product level object

[0055] The basic requirement of fault test isolation design is to make the system have the ability to test and isolate the field level maintainable unit fault through testability design. On the one hand, the fault test design judges the fault related to safety and mission success, and provides the basis for whether to continue to execute the task or maintenance decision; on the other hand, the fault isolation design greatly reduces the fault detection and isolation time in the field repair process, and improves the use availability of the aircraft.

[0056] ​​When designing the function architecture of an aero-engine, the maintainable function units and their functions that need to be designed for testability are identified through function hazard assessment and maintenance level analysis, and then the corresponding test function units are set for the functions of the maintainable function units that need to be tested. For the need of testability modeling, the function types are divided into control execution type, test type and acquisition and self-test type. The control execution type mainly includes engine starting and ignition, fuel supply, adjustable mechanism adjustment, anti-icing bleed air and other control execution functions. The test type mainly includes the function of converting the characteristic parameters such as input / output current, medium and displacement of the control execution function into electrical signals through measurement and sensing. The acquisition and self-test type mainly determines whether the electrical signals collected by the test type function are normal through software logic.

[0057] b. Automatic testability determination value D Oij and independent test determination value D Dij

[0058] Based on mature test technology, the testability of the failure modes of the functions that need to be designed for testability is determined. When the test function can be set for the characteristic parameters such as input / output current, medium and displacement of the control execution function, through comparison of the function input / output characteristic parameter relationship, extreme value detection and other means, and application of state monitoring and fault diagnosis technology based on performance parameter model and historical data model, theoretically all abnormal failure modes of the control execution function can be determined, such as no output, output too large, output too small, output fluctuation, unexpected output, etc. Oij 1. Considering that the function failure characteristic parameter differentiation is not obvious in engineering and other factors cannot effectively determine the failure state, D Oij 0.

[0059] When it is determined that a control execution function of a maintainable function unit can set a test function itself, the independent test determination value D Dij 1, otherwise D Dij 0.

[0060] c. Test function self-test capability value B ij

[0061] When setting a test function, the test function may also be in a failure state. In order to correctly identify that the test function is in a normal state and avoid false alarm, the test function of the corresponding test unit also needs to be self-tested (BIT). The self-test capability needs to be determined based on the existing test technology for the redundancy of the control execution function to be tested, the redundancy of the test function and the input / output characteristic type.

[0062] 1. When the input and output characteristic parameters of the control execution function under test conform to analog characteristics, and the control execution function under test has redundancy or a redundancy test function is set, wherein the redundancy is greater than or equal to 2, assuming there is no common-mode failure, then all typical fault modes of the test function can be self-tested, such as no signal, signal drift, unexpected output signal, etc. B ij Take 1;

[0063] 2. When the input and output characteristic parameters of the control execution function under test conform to analog characteristics, and the control execution function under test has no redundancy or the test function is set to have no redundancy, then only the no-signal and unexpected output signal fault modes related to the test function can be tested, and signal drift fault modes cannot be tested. (B) ij It can be determined by the sum of the percentages of testable fault modes such as no signal and unexpected output signal. If fault mode percentage data is unavailable, B... ij The proportions can be evenly distributed according to the number of testable failure modes;

[0064] 3. When the characteristic parameters of the function being tested conform to the characteristics of a switching quantity, and there is an input control signal, self-testing can be performed on the unexpected output signal fault mode. For the no-signal fault mode, an output signal must be added in the detection program via hardware and software, involving the corresponding load action. Firstly, some loads require hydraulic power when operating, and the engine needs to run to a certain state. Secondly, the operation of some loads will consume the lifespan of accessories, and it is generally not allowed to perform self-testing by detecting the lifespan consumption setting. Therefore, self-testing is generally not performed on the no-signal fault mode. At this time, B ij It can be determined based on the proportion of unexpected output signal failure modes. If failure mode proportion data is unavailable, B... ij The proportion can be evenly distributed according to the number of failure modes, taking 0.5;

[0065] 4. When the characteristic parameters of the function being tested conform to the characteristics of a switch input, and there is no input control signal, self-testing is generally not performed on fault modes such as no signal or unexpected output signal. (B) ij Set to 0;

[0066] 5. When the test function does not have a self-test unit set in the system, B ij Take 0.

[0067] Step S200: Based on the system test architecture data, obtain the probability λ of the fault modes that the i-th basic-level repairable functional unit can be correctly tested. Di The failure rate λ of the i-th basic-level maintainable functional unit i Establish system fault detection rate r FD Mathematical model;

[0068] The system fault detection rate r FD The mathematical model is as follows:

[0069]

[0070] wherein i = 1, 2, 3, …, n, n is the number of base-level maintainable functional units.

[0071] Based on a typical system test architecture model and data, the probability of correct testable failure mode of the ith base-level maintainable functional unit λ Di The mathematical model is:

[0072]

[0073] wherein λ Dij is the probability of correct testable failure mode of the jth control execution type function in the ith base-level maintainable functional unit, j = 1, 2, 3, …, Q Di , Q Di is the number of control execution type functions in the ith base-level maintainable functional unit.

[0074] λ Tik is the probability of correct testable failure mode of the kth test type function in the ith base-level maintainable functional unit, k = 1, 2, 3, …, Q Ti , Q Ti is the number of test type functions in the ith base-level maintainable functional unit.

[0075] λ Bil is the probability of correct testable failure mode of the lth collection and self-test function in the ith base-level maintainable functional unit, l = 1, 2, 3, …, Q Bi , Q Bi is the number of collection and self-test functions in the ith base-level maintainable functional unit. In engineering, ground detection analysis equipment (ATE) is generally used to ensure that the collection and self-test function failure can be 100% detected and isolated.

[0076] When the base-level maintainable functional unit only has test functions in itself, a typical test architecture is shown in the test architecture of the control execution function 1 of the maintainable functional unit 1. Figure 2 The detectable failure probability of a certain control execution function is the product of the function failure mode occurrence probability and the successful detection probability of its test function.

[0077] The mathematical model of the probability of correct testable failure mode of the jth control execution type function in the ith base-level maintainable functional unit λ Dij is:

[0078]

[0079] wherein DOij — The jth control execution class function automatic testability decision value in the ith base-level maintainable function unit;

[0080] α Oij — The proportion of the jth control execution class function failure mode in the ith base-level maintainable function unit. In the architecture design stage, when the failure mode proportion data is missing, the equal division method can be used to divide according to the number of control execution class functions;

[0081] α Dik — The proportion of the kth test function failure mode corresponding to the jth control execution class function in the ith base-level maintainable function unit (when determining the failure mode proportion, the general testability design criterion requirement can be considered, and the test function reliability level should be higher than the tested function reliability level by one order of magnitude, that is, the test function failure probability is not greater than 0.1 times the failure rate of the tested function). In the architecture design stage, since the proportion data is missing, the equal division method can be used to divide according to the number of test class functions;

[0082] λ KBi — The failure rate of the kth test function in the ith base-level maintainable function unit in the acquisition and self-test function unit set in the subordinate upper-level system;

[0083] α KBiq — The proportion of the kth test function failure mode corresponding to the jth control execution class function in the ith base-level maintainable function unit in the acquisition and self-test function unit set in the subordinate upper-level system, q = 1, 2, 3, …, Q Bi , Q Bi is the number of acquisition and self-test class functions in the Bi system-level acquisition and self-test unit;

[0084] R k — The redundancy of the kth test function corresponding to the jth control execution class function in the ith base-level maintainable function unit. Generally, when the test function has redundancy, the acquisition and self-test function has the same redundancy as the corresponding test function.

[0085] t s — The working time of the base-level maintainable function unit in the average mission flight time.

[0086] When the control execution class function of the base-level maintainable function unit completely relies on the corresponding function test unit in the upper-level system to test the function, the typical test architecture is shown in the test architecture of the control execution function 3 of the maintainable function unit 1. Figure 2

[0087] The ith base-level maintainable function unit kth test class function correctable failure mode probability λ Tik ​The mathematical model of is:

[0088]

[0089] Wherein, α Sip The proportion of the pth test function failure mode of the corresponding system-level test unit of the jth control execution class function in the ith basic-level maintainable function unit, p = 1, 2, 3, …, Q Si , Q Si is the number of test class functions in the corresponding system-level test unit of the jth control execution class function.

[0090] λ Si The failure rate of the corresponding function test unit set in the superior system for the testable function in the ith basic-level maintainable function unit;

[0091] λ SBi The failure rate of the acquisition and self-test function unit set in the superior system for the pth test function of the corresponding system-level test unit of the jth control execution class function in the ith basic-level maintainable function unit;

[0092] α SBiq The proportion of the corresponding acquisition and self-test function failure mode of the acquisition and self-test function unit set in the superior system for the pth test function of the corresponding system-level test unit of the jth control execution class function in the ith basic-level maintainable function unit, q = 1, 2, 3, …, Q Bi , Q Bi is the number of acquisition and self-test class functions in the Bi th system-level acquisition and self-test unit;

[0093] R p The margin of the pth test function of the corresponding Si th system-level test unit of the jth control execution class function in the ith basic-level maintainable function unit.

[0094] When the basic-level maintainable function unit sets test functions in itself and the superior system, a typical test architecture is shown in the test architecture of the control execution function 2 of the maintainable function unit 1 in Figure 2 The more test functions and the greater the test function margin of a certain control execution function are set, the higher the probability of correctly detecting its failure is

[0095] The mathematical model of the probability of correctly testable failure mode λ Bil of the lth acquisition and self-test function in the ith basic-level maintainable function unit is:

[0096]

[0097] Wherein, λi Failure rate of the ith basic-level maintainable functional unit;

[0098] B ik The ith basic-level maintainable functional unit, the kth test function self-detection capability decision value. B ik According to the determination method of the test function self-test capability value in step S100.

[0099] Note that the test unit set at the system level due to the need of the system test architecture also belongs to a basic-level maintainable functional unit and has a certain failure rate.

[0100] When the basic-level maintainable functional unit contains test type functions, a typical test architecture is shown in the test architecture of the maintainable functional unit 5 in FIG. 1. Figure 2 The mathematical model of the testable failure mode probability λ Tik of the basic-level maintainable functional unit test type function is:

[0101]

[0102] Wherein, λ i Failure rate of the ith basic-level maintainable functional unit;

[0103] B ik The ith basic-level maintainable functional unit, the kth test function self-detection capability decision value. B ik According to the determination method of the test function self-test capability value in step 1.

[0104] Note that the test unit set at the system level due to the need of the system test architecture also belongs to a basic-level maintainable functional unit and has a certain failure rate.

[0105] When the basic-level maintainable functional unit contains acquisition and self-test functions, a typical test architecture is shown in the test architecture of the maintainable functional unit 5 in FIG. 1. Figure 2 The mathematical model of the testable failure mode probability λ Bil of the basic-level maintainable functional unit acquisition and self-test function type function is:

[0106] λ Bil = B il α Bil λ i P Bil (7)

[0107] Wherein, α Bil The proportion of the lth acquisition and self-test function failure mode in the ith basic-level maintainable functional unit. In the architecture design stage, due to the lack of proportion data, the equal division method can be used according to the number of acquisition and self-test type functions;

[0108] P Bik The test success rate of the i-th item acquisition and self-test function in the i-th base-level maintainable function unit. The acquisition and self-test functions of an aero-engine are generally arranged in one or several electronic controllers with redundancy, and the acquisition and self-test function faults are detected by the detection of the corresponding detection function detection circuit, mainly through software diagnosis logic, and the detection success rate depends on the hardware reliability such as the electronic controller memory and processor and the corresponding software reliability, and the test success probability is generally required to be not less than 99.99%.

[0109] Note that the acquisition and self-test unit arranged at the system level due to the needs of the system test architecture also belongs to a base-level maintainable function unit and has a certain failure rate.

[0110] In step S300, based on the obtained system test architecture data and the failure mode probability λ Di of the function that can be correctly tested by the i-th base-level maintainable function unit, the failure mode probability λ Ii of the function that can be isolated to the i-th base-level maintainable function unit is obtained, and the system fault isolation rate r FI is established.

[0111] The fault isolation rate is calculated based on the fault detection rate and the fault isolation capability of the system to all testable functions. The mathematical model of the system fault isolation rate r FI is:

[0112]

[0113] Wherein, λ Ii is the failure mode probability of the function that can be isolated to the i-th base-level maintainable function unit, i=1, 2, 3, …, n, and n is the number of base-level maintainable function units.

[0114] λ Di is the failure mode probability of the function that can be correctly tested by the i-th base-level maintainable function unit.

[0115] From the typical system test architecture model, the testable functions of the base-level maintainable function unit may include control execution functions and test functions, and the faults thereof belong to the fault isolation range. According to the fault isolation approach, the isolation capability of the test point setting level of all testable functions is mainly considered.

[0116] The mathematical model of the failure mode probability λ Ii of the function that can be isolated to the i-th base-level maintainable function unit is:

[0117]

[0118] Wherein, λIDij The fault mode probability of the jth control execution class function in the ith base level maintainable function unit can be isolated by setting a test function in the base level maintainable function unit. Di Q Di Q is the number of control execution class functions in the ith base level maintainable function unit.

[0119] λ ISij The fault mode probability of the jth control execution class function in the ith base level maintainable function unit can be isolated by setting a test function in the system level.

[0120] λ ITik The fault mode probability of the kth test class function in the ith base level maintainable function unit can be isolated by setting a self-test function in the system level.

[0121] λ IBil The fault mode probability of the lth acquisition and self-test function in the ith base level maintainable function unit can be isolated when the base level maintainable function unit contains acquisition and self-test functions.

[0122] The mathematical model of the fault mode probability λ IDij of the jth control execution class function in the ith base level maintainable function unit by setting a test function in the base level maintainable function unit is:

[0123] λ IDij = D Dij λ Dij (10)

[0124] Wherein, D Dij is the independent testability decision value of the jth control execution class function in the ith base level maintainable function unit.

[0125] λ Dij is the fault mode probability of the jth control execution class function in the ith base level maintainable function unit that can be correctly tested, j = 1, 2, 3, …, Q Di Q Di is the number of control execution class functions in the ith base level maintainable function unit.

[0126] Note: When designing the architecture, the test function unit set in the system level can be integrated in a base level maintainable function unit of a related control execution class function. If the test function tests the overall output characteristics of a related control execution function chain of several base level maintainable units, it cannot be considered as an independent test function set in the base level maintainable function unit.

[0127] When isolation is performed by setting test function units at the system level, the isolation capability is related to the ratio of the failure mode probability of a certain control execution class function in the basic level maintainable function unit to the failure mode probability of the control execution class function associated with the same system test function item in the maintainable function unit included in the function-related system, and the lower the ratio, the lower the fault isolation rate.

[0128] The mathematical model of the failure mode probability λ ISij of the jth control execution class function in the ith basic level maintainable function unit isolated by setting test function units at the system level is:

[0129]

[0130] wherein, nr is the number of maintainable function units associated with the jth control execution class function in the ith basic level maintainable function unit in the function-related system, and nr≤n. nr needs to be obtained based on the same system test function item and the control execution class function analysis of the related system function unit through the test system architecture mapping model.

[0131] r is the number of control execution class functions associated with the jth control execution class function in the ith basic level maintainable function unit in the function-related system, and r≤Q Di . r needs to be obtained through the function mapping matrix of architecture design.

[0132] When the basic level maintenance unit has a test function, the fault isolation capability of the test function is related to the ratio of its failure mode probability in the overall failure mode probability of the self-test loop.

[0133] The mathematical model of the failure mode probability λ ITik of the kth test class function in the ith basic level maintainable function unit isolated by setting self-test function units at the system level is:

[0134]

[0135] wherein, λ Bi is the failure rate of the collection and self-test function of the maintainable function unit set in the upper system corresponding to the kth test function in the ith basic level maintainable function unit;

[0136] α Biq is the ratio of the failure mode of the collection and self-test function in the maintainable function unit set in the upper system corresponding to the kth test function in the ith basic level maintainable function unit, q=1, 2, 3, …, Q Bi , Q BiThe number of acquisition and self-test functions in the ith system-level repairable functional unit.

[0137] Note that the test unit set at the system level due to the need of the system test architecture also belongs to a basic-level repairable functional unit and has a certain failure rate.

[0138] When the basic-level repairable functional unit has acquisition and self-test functions, the fault isolation capability of the functions is related to the proportion of the overall failure probability of the corresponding self-test circuit.

[0139] When the basic-level repairable functional unit contains system acquisition and self-test functions, the fault mode probability λ IBil of the ith acquisition and self-test function in the ith basic-level repairable functional unit can be isolated to

[0140]

[0141] The acquisition and self-test functions of an aero-engine are generally concentrated in one or several redundant electronic controllers. Considering the cost-effectiveness and weight of testability design, a layer of test function will not be set for the self-test circuit to further isolate the test function failure and self-test function. Generally, ground test analysis equipment (ATE) and other methods are used to further isolate the electronic controller from other repairable units containing test functions. For example, in index demonstration, it is stipulated that the electronic controller is manually tested and isolated by ground test analysis equipment (ATE) and other methods. Then,

[0142] λ ITik = λ Tik (14)

[0143] λ IBik = λ Bil (15)

[0144] In step S400, based on the system test architecture data and the function failure mode probability λ Di that the basic-level repairable functional unit can correctly test, the fault mode probability λ Ai that the ith basic-level repairable functional unit with test functions may generate false alarms is obtained, and a mathematical model of the system inherent false alarm rate r FA is established.

[0145] The mathematical model of the system inherent false alarm rate r FA is:

[0146]

[0147] Where λ AiThe fault mode probability of the ith base-level maintainable function unit with a test function that can generate false alarms, i=1, 2, 3, …, n, n being the number of base-level maintainable function units;

[0148] λ Di The fault mode probability of the ith base-level maintainable function unit that can be correctly tested.

[0149] The sources of false alarms mainly include two aspects: one is that the false alarms are caused by immature fault judgment threshold and diagnosis logic, and the other is that the false alarms are caused by test function unit faults. Engineering experience shows that, by relying on the exposure of faults in the whole machine test and test flight process and the maturation process of fault judgment threshold and diagnosis logic, the false alarms caused by immature fault judgment threshold and diagnosis logic can be effectively reduced before finalization. Therefore, in the architecture design, the inherent false alarm rate caused by test function faults in the system architecture should be mainly concerned. According to the definition of false alarm rate, the main false alarm situation is that the base-level maintainable function unit can test the control execution type function normally, the corresponding test function fails, and the system self-test function unit cannot detect the corresponding test function failure.

[0150] The mathematical model of the fault mode probability λ Ai of the ith base-level maintainable function unit with a test function that can generate false alarms is:

[0151]

[0152] Wherein, λ Aik The fault mode probability of the kth test type function of the ith base-level maintainable function unit with a test function that can generate false alarms, k=1, 2, 3, …, Q Ti , Q Ti being the number of test type functions in the ith base-level maintainable function unit.

[0153] When the base-level maintainable function unit has a test function, the mathematical model of the fault mode probability λ Aik of the kth test type function of the ith base-level maintainable function unit with a test function that can generate false alarms is:

[0154] λ Aik = α Dik λ i F BAik (18)

[0155] Wherein, F BAik The self-test failure probability of the kth test function of the ith base-level maintainable function unit, mainly considering the self-test failure probability caused by self-test function faults of the part with self-test capability. When there is no self-test capability or no system self-test function unit is set, i.e. Bik is 0, FBAik is 1.

[0156] The failure probability F of the self-test of the ith base-level maintainable functional unit for the kth test function BAik is:

[0157] F BAik = (1-B ik )+B ik (1-e -tsαKBiqλKBi ) (19)

[0158] wherein λ KBi is the failure rate of the kth test function in the ith base-level maintainable functional unit in the acquisition and self-test functional unit set in the superior system;

[0159] α KBiq is the proportion of the failure mode of the corresponding acquisition and self-test function in the acquisition and self-test functional unit set in the superior system, q = 1, 2, 3, …, Q Bi , Q Bi is the number of acquisition and self-test functions in the Bi th system-level acquisition and self-test unit.

[0160] Step S500, according to the established system failure detection rate r FD mathematical model, the system failure isolation rate r FI mathematical model, the system inherent false alarm rate r FA mathematical model, the system testability prediction is completed, and the system testability prediction includes the system failure detection rate r FD prediction, the system failure isolation rate r FI prediction and the system inherent false alarm rate r FA prediction.

[0161] A specific application example is as follows:

[0162] 1. Determine the functional architecture data

[0163] The base-level maintainable functional unit composition, predicted failure rate, and testability determination of the functions of the fuel and control system of the aero-engine are shown in Table 1. The electronic controller is allowed to be 100% detected and isolated by a ground detection analysis device (ATE). The engine mission time is 1500h.

[0164] Table 1 Fuel and control system test architecture data detail table

[0165]

[0166]

[0167]

[0168] 2. According to the system fault detection rate r FD Mathematical model, predict the system fault detection rate r FD：

[0169] According to the architecture and related data in Table 1, the control execution function, test function, acquisition and self-test function contained in each functional unit are identified, and the correctable fault mode probability of each function is calculated according to formulas (3), (4), (5), (6), (7), respectively. See Table 2 for details.

[0170] Taking the filter fuel (0301) function of the fuel filter as an example, according to the architecture clues in Table 1, the test architecture contains unit self-test function and system test function, which are fuel filter pressure difference alarm (0302) and fuel filter inlet pressure detection (0903), respectively. Fuel filter pressure difference alarm (0302) corresponds to fuel filter pressure difference signal acquisition and self-test (0107) function, and fuel filter outlet pressure detection (0903) corresponds to fuel filter outlet pressure acquisition and self-test (0110). The correctable fault mode probability λ Dij As follows:

[0171]

[0172] Table 2 Fault detection rate calculation details

[0173]

[0174]

[0175] According to formulas (1), (2), the system fault detection rate r FD As follows:

[0176]

[0177] 3. According to the system fault isolation rate r FI Mathematical model, predict the system fault isolation rate r FI :

[0178] According to the architecture and related data in Table 1, the control execution function, test function, acquisition and self-test function contained in each functional unit are identified, and the correctable fault mode probability of each function is calculated according to formulas (10), (11), (12), (13), (14), (15), respectively. See Table 2 for details.

[0179] Take the main fuel control (0102) of the electronic controller as an example, according to the architecture clues in Table 1, this function is related to the ignition control and thrust control systems. This function corresponds to the three system detection function items 0901, 1101, and 1201. The related system with the same detection function item is the thrust control system, which includes the related functions of the main fuel control (0102) and the main fuel (0401) outputting the corresponding pressure flow according to the control law, and the fuel distribution function (0501). The main fuel control function failure mode probability isolated to the electronic controller can be calculated according to formula (11) as follows:

[0180] λ ISij = 0.00027 × ((1-0) × 0.27 × 1000 × 10 -6 ) / ((1-0) × 0.27

[0181] × 1000 × 10 -6 + (1-1) × 0.5 × 200 × 10 -6 + (1-0) × 1

[0182] × 2 × 10 -6 ) = 268.02 × 10 -6

[0183] According to formula (8) and (9), the system fault isolation rate r FI is calculated as follows:

[0184]

[0185] 4. According to the mathematical model of the system inherent false alarm rate r FA , the system inherent false alarm rate r FA is predicted as follows:

[0186] According to Table 1 architecture and related data, identify the test functions contained in each functional unit, calculate the fault mode probability that the test function may generate false alarms according to formula (17) and (18), see Table 2 for details.

[0187] According to formula (16), the system false alarm rate r FA is calculated as follows:

[0188]

[0189] 5. Complete the system testability prediction, obtain the predicted system fault detection rate r FD , system fault isolation rate r FI and system inherent false alarm rate r FA .

[0190] The above disclosed are only specific embodiments of the present application, but the present application is not limited thereto, and any changes that can be thought of by those skilled in the art shall fall within the protection scope of the present application.

Claims

1. A system testability prediction method based on the functional architecture of an aero-engine, characterized in that, Includes the following steps: Based on the functional architecture of the aero-engine, the system test architecture data is determined. This system test architecture data includes the functional types of specified product-level objects and the automatic testability judgment value D. Oij Independent test judgment value D Dij And the self-test capability value B of the testing function ij The functional types include control execution functions, testing functions, and data acquisition and self-testing functions. Based on the system test architecture data, the probability λ of the fault mode that the i-th basic-level repairable functional unit can be correctly tested is obtained. Di The failure rate λ of the i-th basic-level maintainable functional unit i Establish system fault detection rate r FD Mathematical model; Based on the obtained system test architecture data and the probability λ of the fault modes that the i-th basic-level repairable functional unit can be correctly tested, Di Obtain the functional failure mode probability λ that can be isolated down to the i-th basic level maintainable functional unit. Ii Establish system fault isolation rate r FI Mathematical model; Based on the obtained system test architecture data and the probability λ of the functional failure modes that can be correctly tested in the basic-level repairable functional units, Di Obtain the probability λ of the failure mode that may produce a false alarm for the i-th basic-level repairable functional unit with testing function. Ai Establish the system's inherent false alarm rate r FA Mathematical model; Based on the established system fault detection rate r FD Mathematical model, system fault isolation rate r FI Mathematical model, inherent false alarm rate r of the system FA A mathematical model is used to complete the system testability prediction, which includes the system failure detection rate r. FD Expected system fault isolation rate r FI Expected and inherent false alarm rate r FA Expected.

2. The system testability prediction method based on aero-engine functional architecture as described in claim 1, characterized in that, The system fault detection rate r FD The mathematical model is as follows: Where i = 1, 2, 3, ..., n, and n is the number of basic-level maintainable functional units.

3. The system testability prediction method based on aero-engine functional architecture as described in claim 2, characterized in that, The probability λ of the fault mode that can be correctly tested by the i-th basic-level repairable functional unit. Di The mathematical model is as follows: Where, λ Dij —The probability of a correctly testable failure mode for the j-th control execution function within the i-th basic-level maintainable functional unit, j = 1, 2, 3, ..., Q. Di Q Di Let be the number of control execution functions in the i-th basic-level maintainable functional unit; λ Tik —The probability of a correctly testable failure mode for the k-th test class function in the i-th basic-level maintainable functional unit, k = 1, 2, 3, ..., Q. Ti Q Ti The number of test-type functions in the i-th basic-level maintainable functional unit; λ Bil —The probability of a correctly testable fault mode for the l-th data acquisition and self-test function in the i-th basic-level maintainable functional unit, l = 1, 2, 3, ..., Q. Bi Q Bi The number of data acquisition and self-test functions in the i-th basic-level maintainable functional unit.

4. The system testability prediction method based on aero-engine functional architecture as described in claim 3, characterized in that, The probability λ of the correctly testable failure mode of the j-th control execution function in the i-th basic-level maintainable functional unit. Dij include: The probability λ of the correctly testable failure mode of the j-th control execution function in the i-th basic-level maintainable functional unit. Dij , The probability λ of the correctly testable failure mode of the k-th test class function in the i-th basic-level maintainable functional unit. Dij , The probability λ of the correctly testable fault modes of the l-th data acquisition and self-test function in the i-th basic-level maintainable functional unit. Dij , Among them, D Oij —Automatic testability judgment value of the j-th control execution function in the i-th basic-level maintainable functional unit; α Oij —The percentage of the j-th control execution function failure mode in the i-th basic-level maintainable functional unit; α Dik —The percentage of the k-th test function failure modes corresponding to the j-th control execution function in the i-th basic-level maintainable functional unit; λ KBi —The failure rate of the k-th test function in the i-th basic-level maintainable functional unit within the data acquisition and self-test functional unit set in the superior system; α KBiq —The percentage of failure modes of the k-th test function in the i-th basic-level maintainable functional unit within the corresponding data acquisition and self-test function in the superior system; R k —Redundancy of the k-th test function corresponding to the j-th control execution function in the i-th basic-level maintainable functional unit; t s —Working time of basic-level maintainable functional units during mean mission flight time; α Sip —The percentage of the p-th test function failure modes in the system-level test unit corresponding to the j-th control execution function in the i-th basic-level maintainable functional unit; λ Si —The failure rate of the testable functions in the i-th basic-level maintainable functional unit in the corresponding functional test unit set in the superior system; λ SBi —The failure rate of the p-th test function of the system-level test unit corresponding to the j-th control execution function in the i-th basic-level maintainable functional unit in the data acquisition and self-test function unit set in the superior system; α SBiq —The percentage of the failure modes of the p-th test function of the system-level test unit corresponding to the j-th control execution function in the i-th basic-level maintainable functional unit, within the corresponding data acquisition and self-test function in the upper-level system; R p —The redundancy of the p-th test function in the Si-th system-level test unit corresponding to the j-th control execution function in the i-th basic-level maintainable functional unit.

5. The system testability prediction method based on aero-engine functional architecture as described in claim 1, characterized in that, The system fault isolation rate r FI The mathematical model is as follows:

6. The system testability prediction method based on aero-engine functional architecture as described in claim 5, characterized in that, The probability λ of functional failure modes that can be isolated down to the i-th basic level of maintainable functional unit Ii include: By setting up its own testing functions, the failure mode probability λ of the j-th control execution function in the i-th basic-level maintainable functional unit can be isolated. IDij , l IDij =D Dij l Dij ; By setting up test function units at the system level, the failure mode probability λISij of the j-th control execution function in the i-th basic-level maintainable function unit can be isolated. By setting up self-test functional units at the system level, the failure mode probability λ of the k-th test class function in the i-th basic-level maintainable functional unit can be isolated. ITik , When a basic-level maintainable functional unit includes system data acquisition and self-test functions, the failure mode probability λ of the l-th data acquisition and self-test function in the i-th basic-level maintainable functional unit can be isolated. IBil , Where, nr—the number of maintainable functional units associated with the j-th control execution function in the i-th basic-level maintainable functional unit in the function-related system, nr≤n; r—the number of control execution functions associated with the j-th control execution function in the i-th basic-level maintainable functional unit in the function-related system, r≤QDi; λ Bi —The failure rate of the maintainable functional unit corresponding to the k-th test function in the upper-level system set with the data acquisition and self-test functions in the i-th basic-level maintainable functional unit; α Biq —The percentage of fault modes in the data acquisition and self-test functions of the maintenance functional units set in the superior system corresponding to the k-th test function in the i-th basic-level maintenance functional unit, q = 1, 2, 3, ..., Q Bi Q Bi The number of data acquisition and self-test functions in the Bith system-level maintainable functional unit.

7. The system testability prediction method based on aero-engine functional architecture as described in claim 1, characterized in that, The system's inherent false alarm rate r FA The mathematical model is as follows:

8. The system testability prediction method based on aero-engine functional architecture as described in claim 7, characterized in that, The probability λ of the fault mode that may generate a false alarm in the i-th basic-level repairable functional unit with testing function is... Ai The mathematical model is as follows: Where, λ Aik —The probability of a false alarm occurring in the k-th test function of the i-th basic-level maintainable functional unit with testing capabilities, k = 1, 2, 3, ..., Q. Ti Q Ti The number of test functions in the i-th basic-level maintainable functional unit.

9. The system testability prediction method based on aero-engine functional architecture as described in claim 8, characterized in that, When a basic-level maintainable functional unit has a testing function, the probability λ of the fault mode that may generate a false alarm in the k-th test type function of the i-th basic-level maintainable functional unit with testing function is... Aik for: l Aik =a Dik l i F BAik ; Among them, F BAik —The probability of self-test failure of the k-th test function of the i-th basic-level repairable functional unit, mainly considering the probability of self-test failure caused by the failure of the self-test function of the part with self-test capability.

10. The system testability prediction method based on aero-engine functional architecture as described in claim 9, characterized in that, The self-test failure probability F of the k-th test function of the i-th basic-level repairable functional unit is... BAik for:

Images