Gated recurrent unit network based on attention garrote and its application for industrial soft sensors
SUI Lin
MA Jun-xia
XIONG Wei-li
Abstract：The nonlinearity,dynamics and variable redundancy of complex industrial processes lead to increase model-ing difficulty and reduce model performance.Therefore,a gated recurrent unit(GRU)network based on attention mecha-nism and nonnegative garrote(NNG)estimation is proposed and applied to actual industrial process soft sensor modeling.Firstly,the temporal attention is introduced into the GRU network,and the attention weights are adaptively assigned ac-cording to the temporal correlation between the implied layers at different moments to improve the model of temporal feature characterization capability.Secondly,an attentional weight vector for process variables is designed and embed-ded with NNG algorithm constraints to approximate unbiased estimates of its coefficients.Then the NNG algorithm with variable attention is used to perform sparse optimization of the GRU network to reduce model complexity,improve its interpretability and prevent overfitting.The effectiveness and superiority of the algorithm are verified by numerical simu-lation.Finally,the proposed algorithm is applied to the soft sensor of SO2 concentration in net flue gas emissions from a coal-fired power plant desulphurization process.The experimental results show that the proposed algorithm outperforms other advanced comparative algorithms and improves its predictive performance while effectively eliminating redundant variables and simplifying the model structure.
Keywords：soft sensorgated recurrent unitattention mechanismnonnegative garrotevariable selectionsparse optimization
Publication Date：2026-02-28
Online Publishing Date：2026-04-08(First online date of this platform, not the publication date of the document)
Pages：13( 425-437 )