ptimized automatic generation control using single and multiobjective GA and DE techniques
Journal Title: International Journal of Engineering and Science Invention - Year 2018, Vol 7, Issue 9
Abstract
In this paper both single and multi objective genetic algorithm and differential evolution optimization techniques are applied to tune PID parameters for the application of automatic generation control of two area six units non reheat thermal power plant. The controllers are optimized by considering 0.1 step load disturbance for area 1 only and computed with sum of absolute value of ith area control error at time t as objective functions. While employing multi objective GA and DE optimization techniques, best compromise solution of the corresponding PID parameters are obtained based on fuzzy membership function assignment technique. Performance and comparison analysis of GA-PID and DE-PID is done using both single and multi objective optimization techniques and according to the result, the optimized value result obtained by DE is getting better than GA in achieving lesser settling time, undershoot and overshoot in AGC application. Further more, for more realistic and confidential the simulation results obtained from the suggested controllers are compared with that of the simulation obtained from without applying the controller.
Authors and Affiliations
Solomon Feleke, K. Vaisakh
The Comparison of the Medium Domination Number and NonSplit Dominating Set of An Interval Graph
Among the variance application of the theory of domination that have been considered, the one that is perhaps most often discussed concerns a communication net work towards an intervals graph(G) corresponding to an inter...
Microwave Assisted Synthesis and Characterization of O-Alkyl or O-Aryl Trithiophosphate Derivatives of Ruthenium
An environ-economic route is used for the synthesis of ruthenium (III) derivatives of O-alkyl or O-aryl trithiophosphate in 1:3 molar ratio to form Ru[SSK(S)POR]3 where ( R= Me, Et, Pri , Bui , Ph, CH2Ph ) respectively....
An Exploration of Classification Approaches For Real Life Image Sets
Classification is the imperative and challenging task in computer vision. Classification is based on description, texture or similarity of items or things. Image classification refers to the labeling of images into one o...
Current State Of Fibre Rein forced Self-Compacting Concrete
The relative compressive strength of concrete is high and tensile strength is low is a well-known fact, which provoked a host of investigations to circumvent the problems created by the low tensile strength. The presence...
A Study on Eco-friendly Building Management with Respect to Feasibility of Implementation
The purpose of the study is to examine the factors affecting the eco-friendly building management with respect to feasibility of implementation. A cross sectional research on 487 citizens and 500 students through questio...