Newcastle University School of Chemical Engineering and Advanced Materials
INFERENTIAL MEASUREMENT AND CONTROL
CONTENTS
INTRODUCTION
MEASUREMENT PROBLEMS
POPULAR SOLUTIONS
CONCEPTS
TECHNIQUES
IMPLEMENTATION ISSUES
INFERENTIAL CONTROL
BENEFITS
REFERENCES
© COPYRIGHT
MEASUREMENT PROBLEMS

implemnt.gif (7277 bytes)Measurements are used to monitor and ultimately, to control and optimise processes. Automatic control and optimisation procedures enable and facilitate consistent production. However, they require regular and reliable measurements at the appropriate frequency. Difficulties in measuring quality (primary) variables inevitably mean poor or no control at all. Measurement difficulties can be due to a variety of reasons, inclduing:

  • lack of appropriate on-line instrumentation
    Process operation then has to depend on laboratory assays, which means that results can be infrequent and irregular, in addition to long analysis delays. Depending on how the laboratory analyses are carried out, the results may also be subject to reliability problems.
  • reliability of on-line instruments
    On-line sensors may be available but they may suffer from long measurement delays (e.g. gas chromatographs) or may be subject to factors that affect the reliability of the sensor (e.g. drifts and fouling).

In either case, automatic control and optimisation schemes cannot be implemented and as a result, process performance is degraded.

Author: Ming Tham
If you have any comments, please email them  to: ming.tham@ncl.ac.uk

 
Next
Updated: 21 May, 2000

Part of the SWOT Shop site