Simplification and Control of Linear Time Invariant Systems using Evolutionary Techniques

Abstract

In the present work, the application of the evolutionary techniques have newlinebeen explored in two engineering fields: first is for order reduction and second is for newlinecontrol of Linear Time Invariant Systems. newlineFirstly, present research work concerns with the application of Invasive newlineWeed Optimization (IWO) in order reduction. It is a new population based, newlinederivative free numerical stochastic algorithm. It is inspired by biological growth of newlineweed plants. For order reduction, single-input single- output (SISO) continuous time newlinesystems have been considered. The objective function to be minimized is taken as newlineintegral square error (ISE). This error is a difference between the transient responses newlineof higher dimensional system and lower dimensional system. The fitness function is newlineminimized to obtain all coefficients of polynomials of optimized system. The time newlineand frequency responses for step input of system optimized with IWO technique newlinehave been compared with the response of higher dimensional system and of the newlinesystems optimized with other pre researched techniques. Also, the superiority of newlineIWO has been established by comparing parameters in frequency domain, transient newlineresponse s parameters, performance indices in the form of bar charts and tables. All newlinethe comparisons show that the IWO algorithm gives comparably best results as newlinecompared to other previously researched optimizing algorithms. newline