.png){kind=icon}
.png){kind=icon}
Return
Drift-Aware Feature Learning Based on Autoencoder Preprocessing for Soft Sensors
.png){kind=icon}
OA
AI
DOI：10.1002/aisy.202300486.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In this article, a novel approach is presented for drift-aware feature learning aimed at calibrating drift biases in soft sensors for long-term use. The proposed method leverages an autoencoder for data preprocessing to extract expressive signal drift traces features, and incorporates drift characteristics through the latent space representation in a long short-term memory (LSTM) regression neural network. In the results, it is demonstrated that the proposed approach outperforms other typical recurrent neural networks, such as LSTM, gated recurrent unit, and bidirectional LSTM, with a reduced root mean square error of 60% for the training dataset (approximate to 2.5 h) and 80% for the testing dataset (approximate to 20 h). The proposed approach has the potential to optimize the performance of soft sensors with long-term drift and reduce the need for frequent recalibration. By compensating for sensor drift using existing prior information and limited time data, the proposed neural network can effectively reduce the complexity and computational burden of the system, without the need for additional settings or hyperparameter fine-tuning. In this article, a novel approach is presented to address the issue of drift biases in soft sensors for long-term applications, focusing on drift-aware feature learning. An autoencoder is utilized for data preprocessing, enabling the extraction of informative signal drift trace features. In the experimental results, it is indicated that it outperforms the conventional long short-term memory network, achieving a significant 80% reduction in root mean square error when tested on a 20 h dataset.image (c) 2024 WILEY-VCH GmbH
Keywords:
autoencoders
deep learnings
flexible sensors
sensor calibrations
sensor drifts
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
6.1
Papers:
1.9K
Citations:
8.4K