Nonlinear Modeling Method Applied to Prediction of Hot Metal Silicon in the Ironmaking Blast Furnace
Neural networks have been established as versatile tools for nonlinear black-box modeling,but in many data-mining tasks the choice of relevant inputs and network complexity still constitute challenges.Statistical tests for detecting relations between inputs and outputs proposed in the literature are largely based on the theory for linear systems,and laborious retraining make the application of exhaustive search through all possible network configurations impossible but for toy problems.We propose a systematic method based on a pruning algorithm to tackle the problem where an output shall be estimated on the basis of a large set of potential inputs.The modeling method is applied on a complex modeling problem of predicting the silicon content of hot metal produced in a blast furnace.It is demonstrated to find relevant inputs and to yield parsimonious sparsely-connected neural models of the output.
Non-linear black-box modeling Neural networks Structural and parametric optimization Pruning method
NURKKALA Antti PETTERSSON Frank SAX(E)N Henrik
Thermal and Flow Engineering Laboratory,Department of Chemical Engineering,(A)bo Akademi University,Biskopsgatan 8,FI-20500(A)bo,Finland
国际会议
Asia Steel International conference 2012(第五届亚洲钢铁大会)
北京
英文
1-7
2012-09-23（万方平台首次上网日期，不代表论文的发表时间）