CN-122024455-A - Multi-measuring-plate linkage aircraft structure state early warning method

CN122024455ACN 122024455 ACN122024455 ACN 122024455ACN-122024455-A

Abstract

The application belongs to the technical field of electric data processing, and particularly relates to a multi-test-piece linkage airplane structure state early warning method, which classifies each strain piece according to data characteristics, judges a three-level alarm position by applying at least one classification correlation method based on classification results, judges a two-level alarm position by applying a space or structure correlation rule based on the results of the three-level alarm position, judges a first-level alarm position by carrying out clustering judgment on a key area of a structure based on the results of the two-level alarm position.

Inventors

CHANG LIANG
CHEN JIAOJIAO
WAN CHUNHUA
DUAN FENG
WU DAN

Assignees

中国飞机强度研究所

Dates

Publication Date: 20260512
Application Date: 20260410

Claims (9)

1. A multi-test-piece linkage aircraft structure state early warning method is characterized by comprising the following steps: Step S1, strain data of each strain gauge of an aircraft structure static strength test are read in real time, and preprocessing is carried out on the corresponding data; s2, judging a strain gauge with a strain value larger than a preset bad gauge threshold as a bad gauge, and removing the bad gauge; Step S3, classifying each strain gauge according to the characteristics of the strain data; Step S4, based on the classification result of the step S3, determining a three-level alarm part by applying at least one classification association method; s5, based on the result of the three-level alarm part, a space or structure association rule is applied to judge a second-level alarm part; step S6, based on the result of the secondary alarm part, carrying out clustering judgment on a key area of the structure, and determining a primary alarm part; and S7, repeatedly executing the steps S2 to S6 on the strain data of the next batch until all the strain data are processed, and outputting the strain gauge numbers corresponding to the alarms of each stage.
2. The method for pre-warning the structural state of an aircraft with multi-test-piece linkage according to claim 1, wherein the classification of the strain gauge comprises: Invalid data sheet, irregular jitter sheet, strain overrun sheet, total constant fault sheet, segment constant fault sheet, large amplitude oscillation sheet, creep-like sheet, strain rapid increase sheet, high linearity sheet, general linearity sheet, inflection sheet, same direction jump breaking sheet and reverse jump breaking sheet.
3. The method for early warning of the structural state of an aircraft with multi-test-piece linkage according to claim 2, wherein the method for classifying the strain gauge comprises the following steps: invalid data sheet, the absolute value of the strain gauge output value is always smaller than the set low threshold value; an irregular jitter piece, wherein strain is randomly jittered along with time or is not related to the applied load; strain overrun gauge, the absolute value of strain exceeds the preset upper limit threshold; the total fault sheet is characterized in that in the whole test process, the strain output is constant at 0 or a certain fixed constant value; segment constant fault pieces, namely outputting strain to be constant values in a plurality of continuous time segments; The amplitude of strain is larger than a preset amplitude, and the frequency is larger than a preset upper limit frequency; The creep-like sheet is characterized in that after loading, the strain and the load are in linear relation, and after reaching a certain load point, the strain value is kept in a preset range; The slope of the load and strain curve is larger than the set slope at a certain moment after loading; The linear piece is characterized in that strain and load are in linear relation, and the linear fitting degree is higher than a preset high linearity threshold; the strain and the load are in a linear relation, and the linear fitting degree is between a high linearity threshold value and a low linearity threshold value; Inflection sheet, namely inflection occurs on the curve of load and strain; the same-direction jump breaking piece is that the strain is subjected to step mutation, and the jump direction is the same as the data trend direction before jump; reverse jump breaking piece, that is, the strain is suddenly changed in step type, and the jump direction is opposite to the overall trend direction of the data before jump.
4. The method for pre-warning the structural state of an aircraft with multi-test-piece linkage according to claim 3, wherein, In the step S4, the classification association method comprises a flower association method and a threshold value method; the pattern piece association method is that if at least one inflection piece and one strain rapid increasing piece are contained in three strain pieces forming a pattern piece group or two inflection pieces are contained in the three strain pieces, the position of the pattern piece group is defined as a three-level alarm position; The threshold method is that for the inflection piece, the same-direction jump breaking piece and the reverse jump breaking piece, if the strain value of the inflection piece reaches a preset strain threshold value, the position of the strain piece is defined as a three-level alarm position.
5. The method for early warning of a structural state of an aircraft with multi-test-piece linkage according to claim 4, wherein in step S5, the spatial or structural association rule includes a nearest neighbor search method and an inside and outside-side-piece association method; The nearest neighbor searching method comprises the steps of searching N adjacent strain gauges closest to each strain gauge according to preset space coordinates of all the strain gauges, and if M of the N adjacent strain gauges of one strain gauge are defined as three-level alarm positions, newly increasing the position of the strain gauge as a two-level alarm position, wherein N and M are preset positive integers; The inner and outer sheet association method comprises the step of defining a second-level alarm position as a position of a strain sheet pair stuck on the inner and outer sides of the same structural position if both the inner and outer side strain sheets are inflection sheets and the load and the curve of the strain are in opposite inflection directions.
6. The method for warning the structural state of an aircraft with multi-blade linkage according to claim 5, wherein n=5 and m=2.
7. The method for early warning of a state of an aircraft structure with multi-test-piece linkage according to claim 1, wherein in step S6, the clustering determination in the critical area of the structure includes a location-based keyword search method: and if more than P groups of secondary alarm positions appear in the set defined by the same keyword, defining a structural area related to the keyword as a primary alarm position, wherein P is a preset positive integer.
8. The method for early warning of a structural state of an aircraft with multi-blade linkage according to claim 7, wherein the preset keyword comprises at least one of a frame and a rib, and P is more than or equal to 2.
9. The method for early warning of a structural state of an aircraft with multi-test-piece linkage according to claim 3, wherein in step S2, the preset bad-piece threshold is 8000 microstrain.

Description

Multi-measuring-plate linkage aircraft structure state early warning method Technical Field The application belongs to the technical field of electric data processing, and particularly relates to a multi-measuring-plate linkage aircraft structure state early warning method. Background The static strength test of the aircraft structure is a verification means for finally confirming the aircraft structure design and closest to the actual working condition in the ground environment. The test result is a key basis for whether the aircraft can be approved for test flight and put into use. In the test, the displacement and strain response of the structure are measured through a displacement sensor and a strain gauge, and then data such as stress, displacement change and the like are obtained through analysis and are compared with design indexes to judge whether the test piece meets the static strength requirement. Traditional abnormal data identification and state judgment are highly dependent on manual experience. The analyst needs to observe the change rule of the load-strain curve of a large number of strain gauges one by one, so as to screen the abnormality and evaluate the structural state. The method is low in efficiency and high in cost, is greatly influenced by subjective factors of analysts, and is difficult to ensure the interpretation accuracy. Along with the trend of the complexity of the novel aircraft structure, the number of strain gauges for monitoring is increased rapidly, and the urgent requirement of real-time and accurate monitoring on a test site is difficult to be met by the traditional method. In order to improve efficiency, an automatic classification or machine learning algorithm based on feature rules appears in recent years, and quick preliminary screening of the strain gauge data can be realized. However, the methods still have obvious limitations that firstly, the number of the screened abnormal sheets is huge, the spatial relevance is poor, the inherent relevance of the mechanical state of each part of the aircraft structure is difficult to quantitatively evaluate on the whole, and secondly, the multi-source information cannot be effectively fused to form a layered early warning decision, so that the support on the command decision of a test site is weak. On the other hand, if the structure is reinforced based on the anomalies of only a single or a small number of scattered strain gages, the structure weight is increased due to the fact that the design is too conservative. Therefore, an intelligent method capable of integrating multi-test-piece information, quantitatively evaluating structural states and giving graded early warning is needed to solve the contradiction between weak decision support and design conservation in the prior art. Disclosure of Invention In order to solve the above problems, the present application provides a method for early warning of the structural state of an aircraft with multi-test-piece linkage, comprising: Step S1, strain data of each strain gauge of an aircraft structure static strength test are read in real time, and preprocessing is carried out on the corresponding data; s2, judging a strain gauge with a strain value larger than a preset bad gauge threshold as a bad gauge, and removing the bad gauge; Step S3, classifying each strain gauge according to the characteristics of the strain data; Step S4, based on the classification result of the step S3, determining a three-level alarm part by applying at least one classification association method; s5, based on the result of the three-level alarm part, a space or structure association rule is applied to judge a second-level alarm part; step S6, based on the result of the secondary alarm part, carrying out clustering judgment on a key area of the structure, and determining a primary alarm part; and S7, repeatedly executing the steps S2 to S6 on the strain data of the next batch until all the strain data are processed, and outputting the strain gauge numbers corresponding to the alarms of each stage. Preferably, the classification of the strain gage includes: Invalid data sheet, irregular jitter sheet, strain overrun sheet, total constant fault sheet, segment constant fault sheet, large amplitude oscillation sheet, creep-like sheet, strain rapid increase sheet, high linearity sheet, general linearity sheet, inflection sheet, same direction jump breaking sheet and reverse jump breaking sheet. Preferably, the method for classifying the strain gage includes: invalid data sheet, the absolute value of the strain gauge output value is always smaller than the set low threshold value; an irregular jitter piece, wherein strain is randomly jittered along with time or is not related to the applied load; strain overrun gauge, the absolute value of strain exceeds the preset upper limit threshold; the total fault sheet is characterized in that in the whole test process, the strain output is constant at 0 or a certain