Research on Intelligent PID Control Algorithm Based on Neural Network

Lei, Yu; Ma, Jiajia; Liu, Miao; Liu, Panling

doi:10.1109/icitbs53129.2021.00087

Cited by 2 publications

(1 citation statement)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The optimization algorithm is chosen to be Adam. The Adam algorithm can adaptively adjust the learning rate of each parameter, and different learning rates can be used for different parameters, thus making the training more efficient and stable [35].…”

Section: Tadat Model Hyperparameter Settingsmentioning

confidence: 99%

Research on the service condition monitoring method of rolling bearings based on adaptive deconvolution and transformer

Zhang,

Liu,

Wang

et al. 2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

This research suggests a two-channel adaptive deconvolution and Transformer modules (TADAT) based fault diagnostic approach since the feature extraction capabilities of a single-scale model is constrained by the complexity of rolling bearing operating circumstances in actual working settings. First, the original vibration signal is transformed using the fast Fourier transform (FFT) from the time domain to the frequency domain. Next, multi-scale feature information is extracted using two-channel adaptive deconvolution at the beginning of the model. Finally, the global features of the signal are further extracted using tandem combination with Transformer (multi-headed attention), which enhances the model's overall generalization performance. Second, the loss function is improved, and an L2 regularization penalty term and a regularized loss function with label smoothing are implemented. Finally, the model is tested using data from the imbalanced bearing load test bench and the open source dataset. The results show that the model achieves an accuracy of over 98.5% and maintains that accuracy under noisy samples. The test results demonstrate that the rolling bearing service condition monitoring method based on TADAT feature extraction model proposed in this paper has a straightforward model structure, high diagnostic accuracy, and strong generalization performance, providing a useful reference for the rolling bearing service condition monitoring research.

show abstract

Section: Tadat Model Hyperparameter Settingsmentioning

confidence: 99%