Related Experiment Videos
A statistically based filter.
1School of Chemical Engineering, Oklahoma State University, Stillwater 74078-5021, USA. rrr@okstate.edu
ISA Transactions
|June 20, 2002
Summary
A novel statistical process control method effectively filters noise from process variables. This algorithm is demonstrated on a pilot-scale process with significant noise and frequent level changes.
Area of Science:
- Process Engineering
- Data Analysis
- Control Systems
Background:
- Process variables often contain significant noise, complicating analysis and control.
- Traditional filtering methods may struggle with dynamic processes exhibiting frequent changes.
Purpose of the Study:
- To develop a simple and effective noise filtering procedure for process variables.
- To utilize statistical process control (SPC) concepts for enhanced data quality.
Main Methods:
- A new procedure based on SPC principles was developed.
- Algorithm code was implemented for practical application.
- Experimental validation was performed on a pilot-scale process.
Main Results:
- The developed procedure successfully filtered noise from the process variable.
- The method demonstrated effectiveness even with significant and frequent process changes.
- The algorithm's performance was validated in a real-world pilot-scale setting.
Conclusions:
- The SPC-based noise filtering procedure offers a robust solution for improving process data quality.
- This method is particularly suitable for dynamic industrial processes.
- The presented algorithm and experimental results provide a practical tool for process engineers.