Quality monitoring of continuous flow processes
Computers and Industrial Engineering
SAS/ETS User's Guide, Version 6
SAS/ETS User's Guide, Version 6
Time Series Analysis, Forecasting and Control
Time Series Analysis, Forecasting and Control
Time series applied in Romanian economy
AIKED'09 Proceedings of the 8th WSEAS international conference on Artificial intelligence, knowledge engineering and data bases
| Hi-index | 0.00 |
The aim of this work is the parameter estimation of correlated serial disturbance measurements of output data structure from continuous chemical processes. The chemical process is simulated with HYSYS simulator to obtain the dynamic response of outputs when random disturbances implicate the process inputs. Then, stochastic models are applied to take into account these unexplained disturbances, with the scope of monitoring the product quality. As an illustration, the approach is applied to the water concentration output of a Propylene Glycol plant composed by a CSTR (continuous stirred tank reactor) and a distillation unit, when stochastic perturbations are introduced into the feed molar flows. The proposed statistical approach proves to be satisfactory to predict the correlative data structure and in addition, it allows to forecast future variable values after identifying a given data pattern.