CN-121980927-A - Multi-mode collaborative monitoring method

CN121980927ACN 121980927 ACN121980927 ACN 121980927ACN-121980927-A

Abstract

The multi-mode collaborative monitoring method comprises the steps of constructing a geometric reference model of a structural part manufacturing stage to obtain a manufacturing deviation model, acquiring an assembly pose of a product structure to be assembled in real time in an assembly stage, calculating and correcting assembly deviation in the assembly stage process of the product structure to be assembled to complete an assembly correction closed loop to obtain a product structure which is assembled, constructing an assembly geometric state model, acquiring multi-mode service response data of a key monitoring area in the product structure in the use process of the product structure, carrying out multi-mode monitoring on the state of the product structure to form a multi-mode monitoring data set, and obtaining a structural health state vector and damage probability to complete the monitoring method. The invention can be used to achieve continuous expression and interpretable analysis of structural states between manufacturing, assembly and service phases.

Inventors

XIN LIANG
GAO JING
SONG WENLI
ZHANG YOUYI
XIE XUAN
LI DASONG
TANG XIAOJUN
Lang Jinchi
LIU LIXIA
HUI TIANLI
YANG YAODONG
CHEN DONGKANGKANG
YAN ZHENGANG
GAO XIAOSONG

Assignees

北京卫星制造厂有限公司

Dates

Publication Date: 20260505
Application Date: 20251231

Claims (8)

1. The multi-mode collaborative monitoring method is characterized by comprising the following steps of: 1) Construction of geometric reference models for the manufacturing phase of structural elements Performing manufacturing deviation analysis to obtain a manufacturing deviation model ; 2) In the assembly stage, the assembly pose of the product structure to be assembled is obtained in real time And unifies to the same reference coordinate system; 3) During the assembly stage of the product structure to be assembled, the assembly pose obtained in the step 2) is utilized And using the geometric reference model of the manufacturing stage obtained in step 1) Manufacturing deviation model Calculating and correcting the assembly deviation to finish an assembly correction closed loop, obtaining a product structure of which the assembly is finished, and obtaining a corresponding assembly residual error And the real geometric form data after the assembly is completed, and constructing an assembly geometric state model ; 4) Based on geometric reference model Assembly residual Geometric state model for assembly completion Determining an important monitoring area, collecting multi-mode service response data of the important monitoring area in the product structure in the use process of the product structure, and carrying out multi-mode monitoring on the state of the product structure to form a multi-mode monitoring data set; 5) According to manufacturing deviations And assembly residual Geometric reference model Assembling geometric state model Construction of theoretical Strain According to the theoretical strain And step 4) obtaining the multi-mode monitoring data set obtained in the step to obtain the structural health state vector Probability of damage And (5) completing the monitoring method.
2. The multi-modal collaborative monitoring method according to claim 1, wherein the accent service monitoring area in step 4) includes: product structure corresponds assembly residual error Local areas where geometric deviations or residual errors occur.
3. The multi-modal collaborative monitoring method according to claim 2, wherein the critical service monitoring area in step 4) further comprises: Product structure corresponding geometric reference model The position of the stress path and the theoretical stress concentration area; product structure corresponding geometric state model The positions of key assembly interfaces, connection nodes and docking areas; areas of the product structure that are sensitive to environmental loads or fatigue.
4. The multi-modal collaborative monitoring method according to claim 1, wherein the multi-modal monitoring dataset of step 4) includes: Time series of strain The method is used for representing the local deformation response of the structure; Vibration time series For characterizing structural dynamics; temperature time series For characterizing environmental and structural temperature changes.
5. The method of any one of claims 2-4, wherein the structural health vector in step 5) is Comprises theoretical strain Time series of measured strain Time series of measured vibrations And measured temperature time series 。
6. The multi-modal collaborative monitoring method according to claim 5, wherein the probability of impairment in step 5) From theoretical strain And the measured strain time series And (5) determining.
7. The multi-modal collaborative monitoring method according to claim 6, wherein the manufacturing reference frame in step 2) creates an assembly pose From the assembled position Mapping to a manufacturing reference coordinate system Obtaining; Wherein, the In order to rotate the matrix is rotated, Is a translation vector, and is the observation data of the assembly stage.
8. The multi-modal collaborative monitoring method according to claim 6, wherein step 3) assembles a residual error Including translational misalignment in a reference coordinate system And rotational deviation ; Wherein, the And For making reference coordinate system The method comprises the steps of setting an actual rotation matrix and an actual translation vector in a true assembly pose of a product structure to be assembled; And For making reference coordinate system Next, a nominal rotation matrix and a nominal translation vector in a nominal target assembly pose of the product structure to be assembled.

Description

Multi-mode collaborative monitoring method Technical Field The invention relates to the technical field of structural mechanism health monitoring, in particular to a multi-source collaborative monitoring method for a full life cycle of a structural mechanism. The device is widely applied to various large-scale structural mechanisms in the fields of aerospace, constructional engineering, transportation and the like. Background With the continuous development of industrial manufacturing and engineering application, various large-scale structural mechanisms are increasingly widely applied in the fields of aerospace, constructional engineering, transportation and the like. These structures tend to have complex geometries and important functional requirements, thus placing higher demands on their performance monitoring and health assessment at different stages. However, conventional structural monitoring means mostly rely on a single sensor or monitoring mode, such as appearance detection by a visible light camera or measurement of critical parameters by a limited number of sensors. The single monitoring mode often cannot comprehensively capture the health state of the structure, and is difficult to effectively evaluate the manufacturing quality, the assembly precision and the damage or fatigue possibly generated in the service process. Currently, many monitoring methods, while capable of providing local or single structural information, often have dead zones due to the inability to integrate multidimensional data from different sensors, resulting in early failure, microcracking, fatigue or deformation of the structure, which is difficult to discover in time, thereby affecting the long-term safety and reliability of the structure. In addition, existing monitoring systems lack comprehensive, automated, and real-time feedback capabilities in dealing with complex environments, complex structures, and monitoring during long service periods. The state monitoring of the existing large structural member in the manufacturing, assembling and long-term service stages usually adopts mutually independent monitoring and analysis means, such as three-dimensional measurement in the manufacturing stage, pose detection in the assembling stage and strain or vibration monitoring in the service stage. Although the above means have a certain effectiveness at the respective stages, the following problems are still common in practical applications: 1. Lack of uniform geometric references in manufacturing and assembly stages The geometric model obtained in the manufacturing stage is usually only used for quality acceptance and is not used as a spatial semantic reference in the assembly stage, so that assembly pose errors cannot be clearly associated with manufacturing errors, and the definition and adjustment of the assembly errors lack manufacturing basis. 2. The lack of interpretable state association of the assembly phase with the service phase The multi-mode response data such as strain, vibration, temperature and the like collected in the service stage are usually only analyzed based on statistical or empirical thresholds, are difficult to explain in combination with assembly residual errors, and cannot be distinguished whether the response abnormality is caused by manufacturing defects, assembly deviation or service damage. 3. Cross-stage multi-source data lack unified spatiotemporal semantics The vision, pose and multi-mode sensing data acquired at different stages are inconsistent in sampling frequency, time reference and space coordinate system, and the prior art mostly adopts post alignment or local mapping modes, so that reliable fusion is difficult under unified space-time reference. 4. Lack of a priori inference mechanism for manufacturing-assembly-service conditions The manufacturing deviation, the assembly residual error and the service stress response objectively have mechanical causal relation, but the existing monitoring system does not take the manufacturing and assembly state as the prior condition of service analysis, so that the structural health assessment and the service life prediction lack of physical constraint and cross-stage consistency. Therefore, there is a need for a structural full life cycle monitoring method that can establish uniform geometry and state semantics between manufacturing, assembly, and service phases, and interpret and infer service responses based on manufacturing and assembly priors. Disclosure of Invention The invention solves the technical problems of overcoming the defects of the prior art, and provides a multi-mode collaborative monitoring method for the whole life cycle of a structural member, wherein the real geometric state formed in the manufacturing stage is used as a unified space and state prior, and continuous expression and interpretable analysis of the structural state among the manufacturing, assembling and serving stages are realized through manufacturing reference mapp