A deep surrogate model with spatio-temporal awareness for water quality sensor measurement
作者：
Highlights：
• A novel deep surrogate model for estimating water quality variables.
• Encode the temporal and environmental information to improve surrogate accuracy.
• Apply stacked denoising autoencoder to learn water quality features.
• Apply domain adaptation layer to capture the data distribution deviation.
• Evaluate the model with data from a real-world water quality monitoring system.
摘要
•A novel deep surrogate model for estimating water quality variables.•Encode the temporal and environmental information to improve surrogate accuracy.•Apply stacked denoising autoencoder to learn water quality features.•Apply domain adaptation layer to capture the data distribution deviation.•Evaluate the model with data from a real-world water quality monitoring system.
论文关键词：Soft sensor,Deep learning,Semi-supervised learning
论文评审过程：Received 13 July 2020, Revised 5 January 2021, Accepted 12 March 2022, Available online 28 March 2022, Version of Record 5 April 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.116914
