科技创新与应用2025，Vol.15Issue(31)：27-30,4.DOI:10.19981/j.CN23-1581/G3.2025.31.006
基于数据驱动的惯性测试设备倾角回转误差检测
摘要
Abstract
A data-driven machine learning method is proposed to meet the requirement of detecting the spindle inclination rotation error of a dual-axis turntable.A dual-channel laser displacement sensor mounted on the housing is used to collect yaw data of the main shaft in two orthogonal directions in real time,and the average yaw value during one cycle of motion is obtained as an input feature;the inclination rotation error value obtained by traditional detection methods is used as a label to build a data set.The yaw data and target inclination rotation error are normalized according to the known physical upper and lower bounds.Build a multi-layered perceptron model,use the Adam optimizer during the training process,set Epoch to 100,and avoid overfitting with an early stop strategy.Experimental results show that on the training set,MSE≈8×10-6,MAE≈2.2×10-3,and R2≈0.999 9;on the test set,MSE≈1.1×10-5,MAE≈2.6×10-3,and R2≈0.999 8.It is verified that the proposed method can predict the rotation error of the spindle tilt angle without the need for autocollimator.关键词
惯性测试设备/倾角回转误差/数据驱动/人工神经网络/激光位移传感器/双轴转台Key words
inertial test equipment/inclination rotation error/data driving/artificial neural network/laser displacement sensor/dual-axis turntable分类
机械工程引用本文复制引用
黄帅,周卫平,胡斌..基于数据驱动的惯性测试设备倾角回转误差检测[J].科技创新与应用,2025,15(31):27-30,4.基金项目
江西省教育厅科学技术研究项目(GJJ214014) （GJJ214014）
