基于变尺度教与学算法的DCT换挡毂圆度误差评定研究

Evaluation of spatial roundness error based on variable scale teaching and learning algorithms about the DCT shift cam

针对空间圆度误差评定精度较低以及计算速度较慢的问题,提出了一种变尺度教与学算法的空间圆度误差评定方法。首先,基于误差最小区域原则建立了数学模型和目标函数,随后在标准教与学算法的基础上设计变尺度法教与学算法进行优化,提高了算法的收敛速度和计算精度。最后针对某双离合变速器(DCT)换挡毂?12j6轴颈的16个数据进行求解验证,并将计算结果与粒子群算法、改进的蜂群算法和蔡司三坐标测量机结果进行对比。实例表明,变尺度教与学算法与粒子群算法、改进的蜂群算法相比,在空间圆度误差评定上的收敛速度和计算精度均有明显的优势,体现了变尺度教与学算法的优越性。

Aiming at the problems of low accuracy and slow calculation speed in evaluating spatial roundness error,a method of evaluating spatial roundness error based on variable scale teaching and learning algorithm is proposed.Firstly,the mathematical model and objective function are established based on the principle of minimum error region.Then,on the basis of the standard teaching and learning algorithm,the variable-scale teaching and learning algorithm is designed to optimize,which improves the convergence speed and calculation accuracy of the algorithm.Finally,the total 16 spatial roundness measurement data of coordinate measurement about the DCT shift cam(?12j6)are used to solve the problem,and the results are compared with particle swarm optimization and improved bee colony algorithm and Zeiss CMM.The example shows that the calculation accuracy and convergence speed of the variable-scale teaching and learning algorithm are superior to the traditional algorithm in evaluating the spatial roundness error,which embodies the superiority of the algorithm.