【摘要】 为提升回转体内径测量系统的测量精度,解决测量误差补偿问题,提出一种基于梯度提升树(GBDT)的回转体测量系统误差补偿方法。以自主研制的回转体内径测量机为研究对象,首先分析了其主要误差来源,然后测量标准件得到训练样本和测试样本,分别使用训练样本和测试样本对测量系统误差进行建模和补偿,最后与BP神经网络模型进行对比试验。结果表明,基于梯度提升树的测量误差补偿方法具有更好的补偿效果和稳定性,能有效提高测量的精度,使测量误差从4.7μm减小至1.2μm,误差减少了74.5%,具有工程应用价值。更多还原

【Abstract】 In order to improve the measurement accuracy of the rotary body inner diameter measuring machine and solve the measurement error compensation problem, an error compensation method based on gradient boosting decision tree(GBDT) for the rotary body measurement system is proposed. Taking the self-developed rotary inner diameter measuring machine as the research object, the main error sources are analyzed. The training samples and the test samples are obtained by measuring the standard parts. Using training samples to establish an error compensation model based on GBDT. Performance evaluation of the model using test samples. A comparative test was conducted with the BP neural network model. The results show that the measurement error compensation method based on GBDT has better compensation effect and stability, can effectively improve the measurement accuracy. The error is reduced from 4.7μm to 1.2μm,improving the measuring accuracy by more than 74.5%.Therefore,it is valuable for engineering application.更多还原