On-line measurement validation through confidence level based optimal estimation of a process variable
|Author:||Näsi, Jari, Sorsa, Aki|
|Organizations:||University of Oulu, Faculty of Technology, Control Engineering Laboratory
University of Oulu, Faculty of Technology, Control Engineering Laboratory
|Online Access:||PDF Full Text (PDF, 1 MB)|
|Persistent link:|| http://urn.fi/urn:isbn:9514276132
|Publish Date:|| 2004-11-23
In a continuous chemical process, the accuracy and reliability of process and analytical measurements creates the basis for control system performance and ultimately for product uniformity. Measurement results, whether from fast on-line measurement devices or from sample-based laboratory analyses, is the key for selecting the method for process control and analysis. Intelligent and advanced control methods, exploiting measurements, are of no benefit, if the measurements cannot be trusted.
This report presents results from project where target was to develop validation and estimation method for combining real-time redundant signals, consisting of sensor data, and analytical measurements. The validation of on-line measurement uses less frequently updated but more accurate information to validate frequently updated but less accurate on-line measurements. Validation and estimation of measurements are the keys and prerequisites to efficient calibration of on-line measurement device. An estimate of the measured variable is obtained as a weighted average of the on-line measurements and laboratory analyses. The weighting coefficients are recursively updated in real time when new analyse and measurement results are available. The uncertainty of laboratory analysis must be below a specific limit, or the basic assumption of laboratory analyse being reliable and used as reference to on-line measurement device can not be used.
The calibration and estimation filter has been tested by several data sets collected from an operating plant. The calculation of optimal estimate can be used in several industrial applications to produce additional information for more precise process control. In addition, pre-processed data can be used to calculate a "need for maintenance indicator" to warn the operator for sensor breakdowns, wearing or deterioration and detect calibration needs. The operator's workload is thereby reduced especially in problematic situations where measurement and validation signals are not in convergent, by offering calculated best estimation.
Control Engineering Laboratory. Report A
|Copyright information:||This publication is copyrighted. You may download, display and print it for your own personal use. Commercial use is prohibited.|