Research Of On-Machine Verification And Error Feedback Compensation Means For Complex Surface

Parts with complex surfaces have a wide range of applications in mechanical, aerospace and defense industries, and the shape error of the surface has an important effect on the quality and performance of these parts. In practice, however, the machining error of complex surface is unavoidable, it is necessary to sample data on the surface to compensate partial errors, thereby improving the surface quality. Traditional methods such as three coordinates measuring machine(CMM) require additional positioning and clamping, which would introduce the repeat positioning errors and affect the validity of detection data. In this situation, on machine verification(OMV) technology is introduced to solve this problem. The technology integrates processing and measuring together, could achieve sampling and feedback compensation in multi processing stage, which also has great significance in ensuring the machining accuracy and reducing scrap rate. Based on this, this article mainly completed some work of following related aspects:Firstly, the path planning of measurement on complex surface was studied. The traditional traveling salesman problem was changed into an optimal matching problem for analyzing. When the distance matrix was determined, we used the KM algorithm to optimize the detection order of given sampling points, making the total travel segments shortest. Taking into account the adverse effects of differences between the two kinds of problems, and the efficiency of the algorithm itself, division of the measurement region using the idea of clustering was adopted to improve processing efficiency, whilst a kind of link algorithm put forward to fix the incorrect results. In addition, we proposed a static simulation method based on geometrical spatial judgment, which could transform the process of interference checking into some simple geometric intersection, and quickly return the interference area. Then we made corresponding adjustment repeatedly to get an optimal non interference path.Secondly, we put forward the "mirror-match-compensation processing" strategy, using the idea of mirror to eliminate systematic errors caused by inconsistent processing parameters. At the same time, in order to solve the overcut problem in mirroring process, the post compensation process was transformed into a pre compensation process, and particle swarm optimization(PSO) algorithm used to optimize the position and orientation of the surface, minimizing the maximum distance between the mirror points and the surface. Then we conducted the optimization result as a compensation object with updating processing cutter location.Finally, we made the 5-axis OMV process on the DMU 60 mono block five-axis machining center, and completed feedback compensation process with OMV measurement data of complex surface. Analysis of experimental data as well as coordinate measuring experiment verified the effectiveness of the proposed five-axis OMV methods and feedback compensation strategy in this paper.