الفهرس 
AcknowledgementOnly 14 pages are availabe for public view
 |  | from | 117 |  |  |
| | | from | 117 | | |
Abstract
In this thesis it was shown that the step response tuning methods as previ?usly described in earlier chapters are still of the most effective ,yet simple tuning methods.The discussion was mainly focused on studying these tuning methods, how it handles the tuning process. Process models being a core for the step response methods , were introduced and discussed, and assessed for the functionality and adaptability to the process ,the effect of increasing parameters was studied in improving the model accuracy to capture all process data characteristics.A comparative study of other tuning methods was made and a broad classification to these methods was introduced.A common pitfall in step response methods was spotted and discussed. A solution for this pitfall that was introduced and assessed through using multi models concept ,as previously described . Thereafter a digital implementation for it was introduced and discussed.At the end of the thesis a practical verification was done on one of industrial Pill controllers real process in one of the LOBP to completely test the new method.
6.2 Future work•Detailed mathematical analysis of the proposed detection scheme.
•Detailed mathematical analysis ofthe digitization process.
•Extending the new concept to cover wider spectrum of tuning methods.
•Improving the models accuracy to cover faster non industrial process.
•Implementing Model Based Adaptive Tuning like (Lambda, IMC,SIMC.
•Improvements to the Pill like (process output filtering ,enhanced noise filtering).
•Using the Online / Omine. performance assessment to judge the loop performance in a continuous monitoririg fashion.