The estimation of low and high-pass active filter parameters with opposite charged system search algorithm

Abstract

Algorithms are frequently used to solve problems that have a large search space and take a long time to be mathematically solved. They can later be improved with different improvement methods based on the structure and the type of the problem. In this study, the charged system search algorithm (CSS), which has been successfully implemented in the solutions of numerous engineering problems studied within the literature, was improved by introducing opposition-based learning method (OBL) to it in two different methods. With these improved algorithms, solutions were developed for 30-dimensional multimodal test functions in the first place and the results were discussed. In the second place, the parameters of active filters given below were determined from E24 standard series with developed approaches.Filters are electronic circuits that enhance the wanted frequency components of electric signals applied on their inputs and remove harmonics and interferences from these signals. They are divided into two types; active and passive filters. Active filters are produced with transistors or op-amps. They are financially more advantageous compared to passive filters. These filters are preferred especially in low frequencies due to their low costs. Adjustable for a large frequency domain, active filters are very convenient in terms of size and weight and their designs are highly simple. They can easily be connected successively without affecting one another. In this study, the parameter values of Sallen-Key topology Butterworth low and high-pass active filters, which have an extensive area of use, were determined through improved algorithms.My suggestions for the future studies are these: the effects of the opposite position learning approach on other heuristic algorithms can be analysed solving test functions and different engineering problems; different approaches other than the two approaches proposed in this study, can be developed for the opposite position learning concept; LPF and HPF designs solved in the study can be solved for different degrees and stages and finally new designs can be done for different filter types and different resistor and capacitor series that haven’t been handled in this study.