<abstract xmlns="http://www.w3.org/1999/xhtml">
<p>Intelligent control of heavy dielectric coal beneficiation in coal plants is achieved with the help of deep learning models to optimize the control effect. In this paper, through the study of heavy dielectric coal separation methods and processes, a coal separation control optimization strategy based on a radial basis neural network optimized by the ant colony algorithm is proposed, and the RBF network is optimized by clustering using ant colony algorithm, which is used to determine the center and radius of the basic function of the RBF network. The suspension density, ash content of the fine coal and the level of the Hopper bucket, which affect the control effect, are selected as the inputs of the optimized model, and the control strategy is formulated according to the effect after adjusting the parameters. The experimental simulation results show that the ACO-RBF model has less oscillation when the ash value is changed, the final change is smoother, and the root mean square error of the ash value is 0.075%, which is 36.6% less than that of the PID algorithm. With the control strategy optimized by deep learning, the fluctuation range of the level of the qualified media barrel is controlled between 15 and 25 cm, and the volatility pattern of the level is more regular. The control system based on deep learning can better meet the requirements of the coal processing process and effectively improve the efficiency of a coal processing plant.</p>
</abstract>

Intelligent control of heavy dielectric coal beneficiation in coal plants is achieved with the help of deep learning models to optimize the control effect. In this paper, through the study of heavy dielectric coal separation methods and processes, a coal separation control optimization strategy based on a radial basis neural network optimized by the ant colony algorithm is proposed, and the RBF network is optimized by clustering using ant colony algorithm, which is used to determine the center and radius of the basic function of the RBF network. The suspension density, ash content of the fine coal and the level of the Hopper bucket, which affect the control effect, are selected as the inputs of the optimized model, and the control strategy is formulated according to the effect after adjusting the parameters. The experimental simulation results show that the ACO-RBF model has less oscillation when the ash value is changed, the final change is smoother, and the root mean square error of the ash value is 0.075%, which is 36.6% less than that of the PID algorithm. With the control strategy optimized by deep learning, the fluctuation range of the level of the qualified media barrel is controlled between 15 and 25 cm, and the volatility pattern of the level is more regular. The control system based on deep learning can better meet the requirements of the coal processing process and effectively improve the efficiency of a coal processing plant.

Analysis of Intelligent Control Strategy for Heavy Media Coal Separation Process Based on Deep Learning Model

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

Intelligent control of heavy dielectric coal beneficiation in coal plants is achieved with the help of deep learning models to optimize the control effect. In...

Analysis of Intelligent Control Strategy for Heavy Media Coal Separation Process Based on Deep Learning Model

Online veröffentlicht: 27. Sept. 2023

Seitenbereich: -

Eingereicht: 30. Nov. 2022

Akzeptiert: 07. Apr. 2023

DOI: https://doi.org/10.2478/amns.2023.2.00389

Schlüsselwörter
Deep learning, Heavy media coal separation, Control strategy, ACO-RBF model, Ash value, Level fluctuation.

© 2023 Yu Wang et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Analysis of Intelligent Control Strategy for Heavy Media Coal Separation Process Based on Deep Learning Model

Online veröffentlicht: 27. Sept. 2023

Seitenbereich: -

Eingereicht: 30. Nov. 2022

Akzeptiert: 07. Apr. 2023

DOI: https://doi.org/10.2478/amns.2023.2.00389

SchlüsselwörterDeep learning, Heavy media coal separation, Control strategy, ACO-RBF model, Ash value, Level fluctuation.

© 2023 Yu Wang et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Schlüsselwörter
Deep learning, Heavy media coal separation, Control strategy, ACO-RBF model, Ash value, Level fluctuation.