Research on the Application of Taguchi Theory to Verify the Improved Bamboo Forest Growth Optimization Algorithm

This paper presents a study on the application of the Taguchi method to validate and optimize the performance of the Bamboo Forest Growth Optimization Algorithm (BFGO) using the CEC2017 benchmark test functions. Inspired by the growth patterns of bamboo forests in nature, the BFGO algorithm has shown great potential in solving global optimization problems and has emerged as a promising swarm intelligence algorithm. However, several challenges in terms of efficiency, scalability, and inter-group communication strategies still need to be addressed, especially when dealing with large-scale datasets. To overcome these challenges, this study proposes a strategy that incorporates a mutation rate to alleviate the problem of local optima stagnation and employs the Taguchi method to validate the parameter settings for enhancing the efficiency and scalability of the algorithm. The proposed approach is evaluated on the CEC2017 benchmark test functions, which consist of four different types: unimodal, simple multimodal, hybrid, and composition functions. From these four types, a total of 11 functions are selected to test the Taguchi-enhanced Bamboo Forest Growth Optimization Algorithm (TBFGO). The experimental results demonstrate that the proposed method outperforms the original BFGO algorithm in terms of convergence speed, accuracy, and scalability. Furthermore, a comparative analysis is conducted with other swarm intelligence algorithms, including Particle Swarm Optimization (PSO), Whale Optimization Algorithm (WOA), and the original Bamboo Forest Growth Optimization Algorithm. The results indicate that integrating the Taguchi method validation and incorporating the mutation rate strategy contribute to the improved performance of the Bamboo Forest Growth Optimization Algorithm, making it suitable for solving complex function optimization problems. It exhibits significant advantages and competitiveness compared to other swarm intelligence algorithms.