Borehole Positioning Guided by Magnetic Based on DK-MC-WKNN Algorithm
Due to the variable deformations caused by the inferior stiffness of wing ribs,there is a devia-tion between the actual assembly position and the theoretical position of the pre-drilled holes on the wing ribs during the aircraft assembly process.Additionally,the skin covering the ribs and the narrow internal space of the tail wing make it difficult to accurately measure the actual position of the pre-drilled holes in the part coordinate system,which affects the precise positioning of the connection holes.In response to the above-mentioned issues,a drilling-hole method based on magnetic-guided and an improved fingerprint po-sitioning algorithm is proposed.After embedding magnetic sources into pre-drilled holes,the magnetic in-duction strength of the sources is detected from the exterior of the aircraft skin using Hall sensors.The central position of the magnetic sources is calculated by the fingerprint positioning algorithm.To enhance accuracy and efficiency,a DK-MC-WKNN algorithm based on Minimum Classification and Dynamic K-Value is introduced.The improvement strategy involves classifying the minimal feature sources of the sig-nals and automatically adjusting the K-value through a Euclidean distance ratio threshold,which reduces noise and improves the accuracy and efficiency of the algorithm.Experimental results show that,com-pared to the traditional WKNN algorithm,the DK-MC-WKNN algorithm reduces the average error by 12.52％,decreases the root mean square error by 10.8％,improves positioning efficiency by 160％,a-chieves a maximum positioning error of 0.3 mm,and meets the precision requirements for the connection hole locations.
万方数据
维普
