Time-varying Binary Phasmatodea Population Evolution Algorithm

Abstract

Phasmatodea Population Evolution Algorithm (PPE) is an optimization algorithm based on insect behavior. It excels in solving tasks in continuous space. Standard PPE is not suitable for addressing binary problems such as path selection problems, neural network training, and feature selection problems. However, real-world binary problems cannot be solved by the original PPE algorithm. Because binary solutions can only have values of 0 or 1, while the solution space of standard PPE is continuous. To address these issues, we propose the Binary Phasmatodea Population Evolution (BPPE) algorithm and Time-Varying Binary Phasmatodea Population Evolution (TV-BPPE) for dealing with issues with binary properties and study the effect of different transfer functions on the algorithm’s performance.