The paper presents a novel method for crystal structure prediction using particle swarm optimization (PSO) within an evolutionary scheme. This method is designed to predict stable or metastable structures of compounds based on their chemical compositions and external conditions, such as pressure. The approach combines total-energy calculations with PSO to efficiently minimize free energy surfaces. Key features include a geometrical structure factor method to eliminate redundant structures, a variable unit cell size technique to reduce computational costs, and symmetry constraints to enhance convergence and diversity in structure generation. The method has been successfully applied to various systems, including elemental, binary, and ternary compounds, with high success rates, demonstrating its reliability and potential for crystal structure determination. The paper also discusses the implementation details of the PSO algorithm and provides benchmark results for known structures, showing that the method can predict both stable and metastable structures with high efficiency.The paper presents a novel method for crystal structure prediction using particle swarm optimization (PSO) within an evolutionary scheme. This method is designed to predict stable or metastable structures of compounds based on their chemical compositions and external conditions, such as pressure. The approach combines total-energy calculations with PSO to efficiently minimize free energy surfaces. Key features include a geometrical structure factor method to eliminate redundant structures, a variable unit cell size technique to reduce computational costs, and symmetry constraints to enhance convergence and diversity in structure generation. The method has been successfully applied to various systems, including elemental, binary, and ternary compounds, with high success rates, demonstrating its reliability and potential for crystal structure determination. The paper also discusses the implementation details of the PSO algorithm and provides benchmark results for known structures, showing that the method can predict both stable and metastable structures with high efficiency.