Temperature estimation of electric machines using a hybrid model of feed-forward neural and low-order lumped-parameter thermal networks

Temperature estimation of electric machines using a hybrid model of feed-forward neural and low-order lumped-parameter thermal networks
复制标题

使用前馈神经网络和低阶集总参数热网络的混合模型估计电机温度

DOI:
10.1109/iemdc47953.2021.9449548
复制
发表时间:
2021
期刊:
2021 IEEE International Electric Machines & Drives Conference (IEMDC)
影响因子:
--
通讯作者:
J. Böcker
J. Böcker
中科院分区:
--
文献类型:
--
作者:
E. Gedlu;Oliver Wallscheid;J. Böcker

文献摘要

参考文献

被引文献

相似文献

使用集总参数热网络(LPTN)对电机部件进行精确的温度估计需要了解功率损耗和热参数。然而,基于物理方程的建模参数变化模型输入(每个节点的功率损耗和对流热导率),仅使用正常驱动操作期间可用的测量信号作为输入是不实际的。因此,引入了一个基于黑箱前馈神经网络(FFNN)的模型作为低阶LPTN的一部分,该模型使用新引入的逐步系统识别而不是全局识别的经验测量数据集进行参数化。然后,使用三个不同的未见交叉验证数据集评估所提出的混合模型的性能,这些数据集授予平均最大绝对误差为5 K。
Accurate temperature estimation of parts of electric machines using a lumped-parameter thermal network (LPTN) requires knowledge of power loss and thermal parameters. However, modeling parameter varying model inputs (power loss for each node and convective thermal conductances) based on physical equations using only measurement signals available during normal drive operation as inputs is not tangible. Hence, a black-box feed-forward neural network (FFNN) based model is introduced as part of the low-order LPTN which is parametrized using empirical measurements dataset using a newly introduced step-by-step system identification instead of a global identification. Then, the proposed hybrid model performance is evaluated using three different unseen cross-validation data sets, which granted an average maximum absolute error of 5 K.
具有监督机器学习的同步电机中数据驱动的永磁体温度估计:基准
DOI: 10.1109/tec.2021.3052546
发表时间: 2021
影响因子: 4.9
作者:
W. Kirchgässner;O. Wallscheid;J. Böcker
通讯作者: J. Böcker