Hybrid Layered Encoding Cascade Optimization Model in Two-Stage Amplifier Circuit Design

The integration of hybrid computational intelligence techniques into the Layered Encoding Cascade Optimization (LECO) model is driven by a balanced approach, combining exploration through Genetic Algorithm (GA) for broader solutions and focused local search employing Particle Swarm Optimization (PSO). In this model, a distinct external layer introduces a layer encoding structure with multi- resolution capabilities. This structure enables GA functions for integral value individuals, while conducting PSO functions for real-valued individuals as potential solutions. To estimate the efficacy of the GA-PSO LECO model, a specific study involving the design of a two-stage amplifier circuit is carried out. Results indicate that the GA-PSO LECO model surpasses existing APLAC built-in optimization methods, achieving a minimum on S (1,1) and S (2,2) at the same time, simultaneously maximizing dB [S (2,1)] to 16.469 dB from the two-stage amplifier circuit.

I. Introduction

This according to [1], [2], posits that decision analysis in practical contexts scenarios entails identifying the approach and behaviors that best accomplish an aspired objective or goal. Additionally, [3]–[7] discuss how optimization, prevalent in mathematics, computer science, and economics, involves the selection of the optimal element from a set of available alternatives. In essence, optimization entails the systematic determination of real or integer variable values from within a permissible set to minimize or maximize a function. Guanantara (2018) and Kloumann (2020) [8], [9] express their view that utilizing a scalar objective function with actual values is likely the most straightforward formulation.