شماره ركورد كنفرانس :
5041
عنوان مقاله :
Data-Based Soft Sensor for Product Quality prediction in Processes Using Time Varying Parameter Model
Author/Authors :
R. Parvizi Moghadam Center for Process Integration and Control (CPIC) - Department of Chemical Engineering - University of Sistan and Baluchestan, Zahedan, Iran , F. Shahraki Center for Process Integration and Control (CPIC) - Department of Chemical Engineering - University of Sistan and Baluchestan, Zahedan, Iran , J. Sadeghi Center for Process Integration and Control (CPIC) - Department of Chemical Engineering - University of Sistan and Baluchestan, Zahedan, Iran
كليدواژه :
Soft sensor , time varying parameter , Quality estimation , Identification , Data- based modeling , Product quality
عنوان كنفرانس :
The 10th International Chemical Engineering Congress & Exhibition (IChEC 2018)
چكيده فارسي :
چكيده فارسي ندارد.
چكيده لاتين :
The present paper aimed to design and apply a new data-driven soft sensor by time varying parameter (TVP) model. It use for modification of online product quality monitoring and estimation. A dynamic auto regressive exogenous variable (DARX) model has developed to determine the dynamic transfer function with time varying parameters. A simulated distillation column has used for soft sensor performance validation. The result has indicated that final soft sensor model has low complexity and appropriate performance index for column monitoring, even with missing data existence in observed data. Due to high prediction performance, rapid convergence and low complexity of the model, this method can be efficient in industrial processes control and quality prediction improvement.
