Cargando…

A Dual-Attention Recurrent Neural Network Method for Deep Cone Thickener Underflow Concentration Prediction

This paper focuses on the time series prediction problem for underflow concentration of deep cone thickener. It is commonly used in the industrial sedimentation process. In this paper, we introduce a dual attention neural network method to model both spatial and temporal features of the data collect...

Descripción completa

Detalles Bibliográficos
Autores principales: Yuan, Zhaolin, Hu, Jinlong, Wu, Di, Ban, Xiaojuan
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2020
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7085511/
https://www.ncbi.nlm.nih.gov/pubmed/32110906
http://dx.doi.org/10.3390/s20051260
_version_ 1783508948381335552
author Yuan, Zhaolin
Hu, Jinlong
Wu, Di
Ban, Xiaojuan
author_facet Yuan, Zhaolin
Hu, Jinlong
Wu, Di
Ban, Xiaojuan
author_sort Yuan, Zhaolin
collection PubMed
description This paper focuses on the time series prediction problem for underflow concentration of deep cone thickener. It is commonly used in the industrial sedimentation process. In this paper, we introduce a dual attention neural network method to model both spatial and temporal features of the data collected from multiple sensors in the thickener to predict underflow concentration. The concentration is the key factor for future mining process. This model includes encoder and decoder. Their function is to capture spatial and temporal importance separately from input data, and output more accurate prediction. We also consider the domain knowledge in modeling process. Several supplementary constructed features are examined to enhance the final prediction accuracy in addition to the raw data from sensors. To test the feasibility and efficiency of this method, we select an industrial case based on Industrial Internet of Things (IIoT). This Tailings Thickener is from FLSmidth with multiple sensors. The comparative results support this method has favorable prediction accuracy, which is more than 10% lower than other time series prediction models in some common error indices. We also try to interpret our method with additional ablation experiments for different features and attention mechanisms. By employing mean absolute error index to evaluate the models, experimental result reports that enhanced features and dual-attention modules reduce error of fitting ~5% and ~11%, respectively.
format Online
Article
Text
id pubmed-7085511
institution National Center for Biotechnology Information
language English
publishDate 2020
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-70855112020-03-23 A Dual-Attention Recurrent Neural Network Method for Deep Cone Thickener Underflow Concentration Prediction Yuan, Zhaolin Hu, Jinlong Wu, Di Ban, Xiaojuan Sensors (Basel) Article This paper focuses on the time series prediction problem for underflow concentration of deep cone thickener. It is commonly used in the industrial sedimentation process. In this paper, we introduce a dual attention neural network method to model both spatial and temporal features of the data collected from multiple sensors in the thickener to predict underflow concentration. The concentration is the key factor for future mining process. This model includes encoder and decoder. Their function is to capture spatial and temporal importance separately from input data, and output more accurate prediction. We also consider the domain knowledge in modeling process. Several supplementary constructed features are examined to enhance the final prediction accuracy in addition to the raw data from sensors. To test the feasibility and efficiency of this method, we select an industrial case based on Industrial Internet of Things (IIoT). This Tailings Thickener is from FLSmidth with multiple sensors. The comparative results support this method has favorable prediction accuracy, which is more than 10% lower than other time series prediction models in some common error indices. We also try to interpret our method with additional ablation experiments for different features and attention mechanisms. By employing mean absolute error index to evaluate the models, experimental result reports that enhanced features and dual-attention modules reduce error of fitting ~5% and ~11%, respectively. MDPI 2020-02-26 /pmc/articles/PMC7085511/ /pubmed/32110906 http://dx.doi.org/10.3390/s20051260 Text en © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Yuan, Zhaolin
Hu, Jinlong
Wu, Di
Ban, Xiaojuan
A Dual-Attention Recurrent Neural Network Method for Deep Cone Thickener Underflow Concentration Prediction
title A Dual-Attention Recurrent Neural Network Method for Deep Cone Thickener Underflow Concentration Prediction
title_full A Dual-Attention Recurrent Neural Network Method for Deep Cone Thickener Underflow Concentration Prediction
title_fullStr A Dual-Attention Recurrent Neural Network Method for Deep Cone Thickener Underflow Concentration Prediction
title_full_unstemmed A Dual-Attention Recurrent Neural Network Method for Deep Cone Thickener Underflow Concentration Prediction
title_short A Dual-Attention Recurrent Neural Network Method for Deep Cone Thickener Underflow Concentration Prediction
title_sort dual-attention recurrent neural network method for deep cone thickener underflow concentration prediction
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7085511/
https://www.ncbi.nlm.nih.gov/pubmed/32110906
http://dx.doi.org/10.3390/s20051260
work_keys_str_mv AT yuanzhaolin adualattentionrecurrentneuralnetworkmethodfordeepconethickenerunderflowconcentrationprediction
AT hujinlong adualattentionrecurrentneuralnetworkmethodfordeepconethickenerunderflowconcentrationprediction
AT wudi adualattentionrecurrentneuralnetworkmethodfordeepconethickenerunderflowconcentrationprediction
AT banxiaojuan adualattentionrecurrentneuralnetworkmethodfordeepconethickenerunderflowconcentrationprediction
AT yuanzhaolin dualattentionrecurrentneuralnetworkmethodfordeepconethickenerunderflowconcentrationprediction
AT hujinlong dualattentionrecurrentneuralnetworkmethodfordeepconethickenerunderflowconcentrationprediction
AT wudi dualattentionrecurrentneuralnetworkmethodfordeepconethickenerunderflowconcentrationprediction
AT banxiaojuan dualattentionrecurrentneuralnetworkmethodfordeepconethickenerunderflowconcentrationprediction