An Improved Growth Optimizer and Its Simulation
Shanshan Luo, Qiang Liu · 2025
Addressing the problems of slow convergence speed of growth optimizer, low optimization accuracy, easy to fall into local optimum, and difficult to be well balanced for exploration and exploitation, an improved growth optimizer is proposed. This innovative method enhances the initial population quality and improves the population diversity by applying the Latin hypercube method in the initialization stage of the population. It also dynamically adjusts the upper echelon of society and employs a nonlinear optimization strategy to achieve an effective balance between exploration and exploitation capabilities, which further improves the convergence speed of the algorithm and its ability to jump out of the local optimum. Simulation experiments on several types of standard test functions and two engineering application problems show that the algorithm has high optimization performance and robustness.