Gear flank modification is essential to reduce the noise generated in the gear meshing process,improve the gear transmission performance,and reduce the meshing impact.Aiming at the problem of solving the additional mo...Gear flank modification is essential to reduce the noise generated in the gear meshing process,improve the gear transmission performance,and reduce the meshing impact.Aiming at the problem of solving the additional motions of each axis in the higher-order topology modification technique and how to accurately add the different movements expressed in the form of higher-order polynomials to the corresponding motion axes of the machine tool,a flexible higher-order gear topology modification technique based on an electronic gearbox is proposed.Firstly,a two-parameter topology gear surface equation and a grinding model of wheel grinding gears are established,and the axial feed and tangential feed are expressed in a fifth-order polynomial formula.Secondly,the polynomial coefficients are solved according to the characteristics of the point contact when grinding gears.Finally,an improved electronic gearbox model is constructed by combining the polynomial interpolation function to achieve gear topology modification.The validity and feasibility of the modification method based on the electronic gearbox are verified by experimental examples,which is of great significance for the machining of modification gears based on the continuous generative grinding method of the worm grinding wheel.展开更多
为了提高利用监控和数据采集(supervisory control and data acquisition,SCADA)多变量长时间序列预测齿轮箱油温的精度,解决不同风电机组因处不同运行环境导致的数据分布不一致的问题,提出了一种基于多分支时间序列预测与迁移学习相结...为了提高利用监控和数据采集(supervisory control and data acquisition,SCADA)多变量长时间序列预测齿轮箱油温的精度,解决不同风电机组因处不同运行环境导致的数据分布不一致的问题,提出了一种基于多分支时间序列预测与迁移学习相结合的齿轮箱状态监测方法。首先,利用极致梯度提升(extreme gradient boosting,XGBoost)算法筛选输入参数组成原始序列,对其进行分解得到季节与趋势序列。其次,提出季节、趋势序列特征提取模块获取季节及趋势特征的序列,将其与经过Informer模型处理后的特征序列进行融合后输入进多层感知机映射成最终的预测值,以构建提出的多分支时间序列预测网络(multi-branch time series prediction network,MBFN)。最后,利用迁移学习并结合一分类向量支持机(one-class support vector machine,OCSVM)模型及滑动窗口构建齿轮箱的健康指数,完成齿轮箱状态监测。实验结果表明,所提出模型的MBFN显著提高了油温预测精度,优于常规时间序列预测模型,所使用的迁移策略能以较少数据适应不同数据的分布,进而实现对齿轮箱的状态监测,并且所提出的模型可以提前18.9 d发出齿轮箱故障预警。展开更多
基金Projects(52275483,52075142,U22B2084)supported by the National Natural Science Foundation of ChinaProject(JZ2023HGPA0292)supported by the Fundamental Research Funds for the Central Universities of China。
文摘Gear flank modification is essential to reduce the noise generated in the gear meshing process,improve the gear transmission performance,and reduce the meshing impact.Aiming at the problem of solving the additional motions of each axis in the higher-order topology modification technique and how to accurately add the different movements expressed in the form of higher-order polynomials to the corresponding motion axes of the machine tool,a flexible higher-order gear topology modification technique based on an electronic gearbox is proposed.Firstly,a two-parameter topology gear surface equation and a grinding model of wheel grinding gears are established,and the axial feed and tangential feed are expressed in a fifth-order polynomial formula.Secondly,the polynomial coefficients are solved according to the characteristics of the point contact when grinding gears.Finally,an improved electronic gearbox model is constructed by combining the polynomial interpolation function to achieve gear topology modification.The validity and feasibility of the modification method based on the electronic gearbox are verified by experimental examples,which is of great significance for the machining of modification gears based on the continuous generative grinding method of the worm grinding wheel.
文摘为了提高利用监控和数据采集(supervisory control and data acquisition,SCADA)多变量长时间序列预测齿轮箱油温的精度,解决不同风电机组因处不同运行环境导致的数据分布不一致的问题,提出了一种基于多分支时间序列预测与迁移学习相结合的齿轮箱状态监测方法。首先,利用极致梯度提升(extreme gradient boosting,XGBoost)算法筛选输入参数组成原始序列,对其进行分解得到季节与趋势序列。其次,提出季节、趋势序列特征提取模块获取季节及趋势特征的序列,将其与经过Informer模型处理后的特征序列进行融合后输入进多层感知机映射成最终的预测值,以构建提出的多分支时间序列预测网络(multi-branch time series prediction network,MBFN)。最后,利用迁移学习并结合一分类向量支持机(one-class support vector machine,OCSVM)模型及滑动窗口构建齿轮箱的健康指数,完成齿轮箱状态监测。实验结果表明,所提出模型的MBFN显著提高了油温预测精度,优于常规时间序列预测模型,所使用的迁移策略能以较少数据适应不同数据的分布,进而实现对齿轮箱的状态监测,并且所提出的模型可以提前18.9 d发出齿轮箱故障预警。
