1. Optimization of aluminum fluoride addition in aluminum electrolysis process based on pruned sparse fuzzy neural network.
- Author
-
Wang, Jie, Xie, Yongfang, Xie, Shiwen, and Chen, Xiaofang
- Subjects
ELECTROLYSIS ,GAUSSIAN mixture models ,ALUMINUM ,FUZZY neural networks ,FLUORIDES ,HUMAN fingerprints - Abstract
The aluminum fluoride (AF) addition in aluminum electrolysis process (AEP) can directly influence the current efficiency, energy consumption, and stability of the process. This paper proposes an optimization scheme for AF addition based on pruned sparse fuzzy neural network (PSFNN), aiming at providing an optimal AF addition for aluminum electrolysis cell under normal superheat degree (SD) condition. Firstly, a Gaussian mixture model (GMM) is introduced to identify SD conditions in which the operating modes of AEP are unknown. Then, PSFNN is proposed to establish the AF addition model under normal SD condition identified by GMM. Specifically, a sparse regularization term is designed in loss function of PSFNN to extract the sparse representation from nonlinear process data. A structure optimization strategy based on enhanced optimal brain surgeon (EOBS) algorithm is proposed to prune redundant neurons in the rule layer. Mini-batch gradient descent and AdaBound optimizer are then introduced to optimize the parameters of PSFNN. Finally, the performance is confirmed on the simulated Tennessee Eastman process (TEP) and real-world AEP. Experimental results demonstrate that the proposed scheme provides a satisfactory performance. • A novel pruned sparse fuzzy neural network (PSFNN) is developed to establish the AF addition model under normal SD condition. • A sparse regularization term based on Kullback–Leibler divergence is designed in loss function of PSFNN to obtain the sparse representation. • The parameters of network are learned based on mini-batch gradient descent method with AdaBound optimizer. • An enhanced optimal brain surgeon (EOBS) algorithm is proposed to obtain a compact network structure. • The experimental results demonstrated that our method achieves a satisfactory performance. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF