Prediction Method Of Assembly Gap Based On Point-Cloud Reconstruction For Aeronautical Parts

In the assembly process of large-scale aeronautical components,assembly gaps are easily generated at the assembly joints of parts due to manufacturing variations and assembly errors.In order to ensure the quality of the assembly,it is necessary to effectively measure the assembly gap to guide the gap filling.Traditional measurement methods of aerospace parts gap include feeler gauge detection and manual observation.The measurement efficiency is low,the measurement data is limited,and the artificial influence is large.It is difficult to meet the requirements of high quality and high efficiency of digital assembly.Therefore,researching a new digital gap measurement technique for measurement and analysis of assembly gaps is critical to the improvement of the quality of large-scale aeronautical digital assembly.In this paper,the related technology of aerospace parts assembly gap evaluation is researched,and a method for evaluating the clearance of aviation parts based on point cloud reconstruction is proposed.The structure design and circuit integration of laser-assisted binocular measurement system are carried out.The main research contents are as follows:(1)In order to remove different kinds of noise quickly and accurately in large-field industrial environment,this paper proposes a classification denoising method for multi-scale noise.Aiming at large-scale isolated noise,a median denoising based on region segmentation is proposed.In this way,large-scale noise is removed by setting the median threshold in subregions.A fast bilateral filtering method based on threshold segmentation is proposed for small-scale redundant noise,and the small-scale noise of the boundary is removed.Compared with the traditional denoising method,the accuracy is improved by more than 16%,which can meet the measurement requirements of field test.(2)Aiming at the existing point cloud feature extraction method,this paper proposes a point cloud feature extraction method based on multi-criteria parameters for the problem that the extraction effect of different kinds of feature points is not obvious.According to the measurement requirements of large aviation parts,the point cloud data feature points are reclassified,corresponding criterion parameters are set for different kinds of feature points,and the feature judgment function is established.Compared with several common extraction methods,the extraction effect is obvious and the feature information is complete.(3)A gap prediction method based on point cloud reconstruction is proposed.On the basis of theoretical digital model,the point cloud data of two assembly parts are virtual assembled,the assembly deviation is measured,the deviation fusion is completed,and the assembly gap trend function is calculated.The gap prediction function is assembled to evaluate the assembly gap.Compared with the measurement results at the key points,the gap change trend is basically consistent with the actual situation.(4)Based on the above research,according to the size and measurement requirements of the relevant measuring equipment,the shape and internal structure of the measuring system are designed,the connection scheme of the line is studied,the hardware integration scheme of the measuring system is designed,and the construction and integration of the measuring system is completed..Experiments were carried out at the aviation industry site with a typical composite wing box segment and an aircraft wing as research objects.By comparing with key point measurements,the experimental results show that the maximum error is 0.08 mm and the average error is 0.054 mm.This proves the effectiveness of the proposed gap evaluation method.