Please use this identifier to cite or link to this item:
Title: Computational intelligent techniques for design of model predictive controller in continuous stirred tank reactor
Researcher: Baranilingesan I
Guide(s): Deepa S N
Keywords: Engineering and Technology
Engineering Electrical and Electronic
Model predictive controller
Continuous Stirred Tank Reactor Process
University: Anna University
Completed Date: 2019
Abstract: In chemical industries the Continuous Stirred Tank Reactor CSTR plays an important role to achieve the increased production rate with respect to their corresponding raw materials It is noted that the CSTR process is a non-linear one and in order to increase the production rate; the wastage is to be minimized with possible control actions The reaction involved in CSTR process is associated with energy release or energy absorbed from its surroundings because of these exothermic or endothermic reactions it is necessary to maintain the temperature of the process by employing specific efficient control strategies Based on the exothermic or endothermic reactions of CSTR process the temperature changes greatly influence the reaction rate and henceforth the final product gets altered To maintain the concentration of the reactants in the CSTR it is necessary to vary the temperature rate of the process which is accomplished by circulating coolant around the jacketed CSTR The proposed research contribution in this thesis aims to design an effective controller for modeling the system dynamics and to maintain the product at desired concentration Intelligent PID (Proportional-Integral-Derivative) controller is designed for the considered system since the adopted conventional methodology does not meet the required performance criteria newline The classical optimization strategies such as quadratic programming gradient based method linear programming and interior point method; possess unavoidable limitations of extended execution time complex local optima and global optima occurrences To improve the performance of the PID controller effective evolutionary based optimization algorithms are proposed Instead of conventional methods adopted for tuning the controller parameters evolutionary algorithms such as Particle Swarm Optimization PSO Gravitational Search Algorithm GSA Deterministic Particle Swarm Optimization DPSO and Differential Gravitational Search Algorithm DGSA are proposed here The effective characteristics of the algorithms are combined as hybrid optimization algorithm; in the proposed work hybrid PSO-GSA and hybrid DPSO-DGSA are developed and successfully employed for controller tuning newline newline
Pagination: xxxiii, 224p.
Appears in Departments:Faculty of Electrical Engineering

Files in This Item:
File Description SizeFormat 
01_title.pdfAttached File131.74 kBAdobe PDFView/Open
02_certificates.pdf694.27 kBAdobe PDFView/Open
03_abstracts.pdf207.67 kBAdobe PDFView/Open
04_acknowledgements.pdf309.73 kBAdobe PDFView/Open
05_contents.pdf297.91 kBAdobe PDFView/Open
06_listoftables.pdf224.37 kBAdobe PDFView/Open
07_listoffigures.pdf237.8 kBAdobe PDFView/Open
08_listofabbreviations.pdf350.12 kBAdobe PDFView/Open
09_chapter1.pdf304 kBAdobe PDFView/Open
10_chapter2.pdf368.16 kBAdobe PDFView/Open
11_chapter3.pdf527.59 kBAdobe PDFView/Open
12_chapter4.pdf726.37 kBAdobe PDFView/Open
13_chapter5.pdf698.98 kBAdobe PDFView/Open
14_chapter6.pdf480.06 kBAdobe PDFView/Open
15_chapter7.pdf485.82 kBAdobe PDFView/Open
16_chapter8.pdf506.19 kBAdobe PDFView/Open
17_conclusion.pdf140.25 kBAdobe PDFView/Open
18_appendices.pdf269.16 kBAdobe PDFView/Open
19_references.pdf259.04 kBAdobe PDFView/Open
20_listofpublications.pdf177.74 kBAdobe PDFView/Open
80_recommendation.pdf154.97 kBAdobe PDFView/Open

Items in Shodhganga are protected by copyright, with all rights reserved, unless otherwise indicated.

Altmetric Badge: