深孔直线度检测系统设计与评定研究

发布时间：2018-03-08 02:20

本文选题：深孔直线度　切入点：随机数点精度逼近法　出处：《中北大学》2017年硕士论文　论文类型：学位论文

【摘要】：深孔零件是机械工程领域不可缺少的重要零件,如枪炮管、液压缸、气压缸、动车主轴等。深孔直线度是评价深孔零件合格与否的重要指标;同时,深孔直线度的检测理论及方法与检测结果息息相关。在多数情况下,必须检测深孔直线度。由于深孔存在螺旋沟槽、圆度误差、孔内表面质量误差、轴线及其直线度评定基线难以确定等问题导致深孔直线度很难准确测算。为快速、准确检测出深孔直线度,本文经过对现有研究结果的分析,提出了深孔直线度检测方案与评定算法,设计了检测装置,并编写了相应的程序以实现深孔直线度的检测与评定。首先,设计了利用超声测厚仪的检测装置和利用激光、PSD与三轴加速度传感器的检测装置;提出了利用多圆心替代法、最速下降法与最小二乘法确定深孔轴线的思想,并将其应用于相应的检测系统中。其次,提出了满足最小条件的随机数点精度逼近法与正方网格迭代寻优评定深孔直线的方案;通过对同一组检测数据的评定,验证了本文算法的优越性。较其他评定方法,本文算法在一定程度上达到了提高评定结果准确性和精度的目的。然后,编写了随机数点精度逼近法与正方网格迭代寻优的MATLAB计算程序,实现了上述算法的工程应用;此外,利用本文中的评定算法对同一条空间直线的相同采样数据进行直线度评定,将评定结果与其他算法的评定结果进行对比,以判断本文算法的优劣。结果显示,随机数点精度逼近法与正方网格迭代寻优的评定精度均有提高。此外,正方网格迭代寻优评定直线度精度更高,而随机数点精度逼近法计算速度更快、评定效率更高。
[Abstract]:Deep-hole parts are indispensable parts in the field of mechanical engineering, such as gun tube, hydraulic cylinder, pneumatic cylinder, motorcar spindle, etc. The straightness of deep hole is an important index to evaluate the eligibility of deep-hole parts. The detection theory and method of deep hole straightness are closely related to the test results. In most cases, it is necessary to detect the deep hole straightness. Because there are spiral grooves, roundness errors and inner surface quality errors in deep holes, It is very difficult to calculate the straightness of deep hole because of the difficulty of determining the baseline of axis and its straightness evaluation. In order to detect the straightness of deep hole quickly and accurately, this paper analyzes the existing research results. The detection scheme and evaluation algorithm of deep hole straightness are put forward, the detecting device is designed, and the corresponding program is written to realize the detection and evaluation of deep hole straightness. The detection device using ultrasonic thickness measuring instrument and using laser PSD and triaxial acceleration sensor is designed, and the idea of determining the axis of deep hole by using multi-center substitution method, the steepest descent method and the least square method is put forward. And it is applied to the corresponding detection system. Secondly, the method of random number accuracy approximation and square grid iterative optimization is proposed to evaluate the deep hole straight line, and the evaluation of the same set of detection data is made. The superiority of this algorithm is verified. Compared with other evaluation methods, the algorithm achieves the purpose of improving the accuracy and accuracy of the evaluation results to a certain extent. The MATLAB program of random number accuracy approximation method and square grid iterative optimization is compiled to realize the engineering application of the above algorithm. In addition, the straightness of the same sampling data of the same spatial straight line is evaluated by using the evaluation algorithm in this paper. The evaluation results are compared with those of other algorithms to judge the merits and demerits of this algorithm. The results show that the evaluation accuracy of the random number accuracy approximation method and the square grid iterative optimization method are both improved. The accuracy of straightness evaluation is higher than that of square grid iterative optimization, while the accuracy approximation method of random number points is faster and the evaluation efficiency is higher.
【学位授予单位】：中北大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TG83

【参考文献】