智能系统学报2025，Vol.20Issue(6)：1366-1378,13.DOI:10.11992/tis.202502004
基于非负绞杀的稀疏化ONLSTM及其工业软测量建模
Sparse ONLSTM and non-negative constrained industrial soft sensing modeling
摘要
Abstract
Industrial processes often exhibit characteristics such as multivariable interactions,nonlinear behaviors,and time-varying changes.Thus,the resulting modeling data contain excessive redundant information and complex time-de-pendent patterns,which increase modeling complexity and degrade the model performance.To address these challenges,an ordered neurons long short-term memory(ONLSTM)network integrated with non-negative garrote-based regulariza-tion is proposed herein for industrial soft sensor modeling.First,the shrinkage coefficients of the non-negative garrote are embedded into the weight matrix of the ONLSTM input layer to eliminate redundant input nodes and enable vari-able selection.Second,these coefficients are integrated into the weight matrix of the ONLSTM hidden layer to assign weights based on the importance of hidden neurons.Consequently,redundant nodes and their corresponding informa-tion pathways are pruned,achieving sparse optimization of the network structure.The proposed method is validated via numerical simulations and subsequently employed to predict the SO2 concentration in flue gas emissions from a desul-furization process in a thermal power plant.Experimental results demonstrate that the algorithm effectively selects vari-ables and sparsely optimizes the model structure while maintaining high predictive performance,offering promising pro-spects for broader industrial applications.关键词
软测量/长短时记忆网络/有序神经元/非负绞杀/冗余信息/变量选择/稀疏优化/深度学习Key words
soft sensor/long short-term memory networks/ordered neurons/non-negative strangulation/redundant in-formation/variable selection/sparse optimization/deep learning分类
信息技术与安全科学引用本文复制引用
GUO Yingchen,SUI Lin,XIONG Weili..基于非负绞杀的稀疏化ONLSTM及其工业软测量建模[J].智能系统学报,2025,20(6):1366-1378,13.基金项目
国家自然科学基金项目(61773182) （61773182）
江南大学"轻工技术与工程"双一流学科与支撑学科协同发展支持计划(QGJC20230203). （QGJC20230203）
